Intel® Xeon® processor E7-8800/4800 v3 Application Showcase

Real-TimeInsights.RealBusinessAdvantages.
Intel® Xeon® ProcessorE7v3FamilyApplicationShowcase
LeadershipPerformance.Scalability.World-ClassReliability&Uptime Oct. 2015
1

Intel®Xeon®processorE7v3Family Applicationshowcase
TableofContents
2
 Software Ecosystem………………………………………Slide 3
 Application Showcase Summary………………….Slide 6
 What’s Inside?.............................................................Slide 6
 Platform Value............................................................Slide 7
 Operating System Optimizations…………………Slide 15
 Proof Points by Segment; indicates new since May 2015:
 Big Data Analytics…………………………………….Slide 20
 Core Business & Telecommunications…..Slide 25
 Enterprise Database……………………...………...Slide 39
 Financial Services………………………...…………..Slide 50
 Health & Life Sciences……………………....……..Slide 54
 Technical Computing……………………………….Slide 61
 Better Together and 2x2 Proof Points…………Slide 66
 Helpful Links…………………………………………………..Slide 81
 Configurations & Disclaimers………………………..Slide 83
Use the horizontal slide bar (preview) below to move quickly to a specific section
N E W

* Other Names and Brands May Be Claimed as the Property of Others 3
SoftwarePartnersEnablingontheIntel® Xeon® processorE7v3Family
INTEL CONFIDENTIAL – USE UNDER NDA ONLY

Enterprise
Database
Altibase*
AsiaInfo*
Baosight*
Exasol*
IBM Software*
Kingbase*
MariaDB*
Nomura Research
Institute*
Oracle*
Oracle MySQL*
SAP*
Core Business &
Telecommunications
Digital China*
Huawei*
IBM Software*
MemSQL*
Microsoft*
Neusoft*
Oracle*
Red Hat*
SUSE*
TongTech*
TYDIC*
VMware*
Yonyou*
Health &
Life Sciences
Dassault Systemes*
(BIOVIA*)
Epic*
GROMACS*
InterSystems*
Scripps
Translational
Institute*
Vital Images*
Financial
Services
Kingdom
Technology*
Kx*
SunGard*
Big Data
Analytics
Ayasdi*
NARI Technology
Development Co.,
Ltd.*
SAS*
Software AG /
Apama*
Technical
Computing
Altair*
Ansys*
Mentor Graphics*
By Business Segment

* Other Names and Brands May Be Claimed as the Property of Others
ApplicationshowcaseSummary
This set of Intel® Xeon® processor E7-8800/4800 v3 family proof points spans several key business segments. They represent the significantly
better generation-to-generation processor performance seen with the Intel Xeon processor E7-8800/4800 v3, which is further enhanced by
using other Intel ingredients such as the Intel® Solid-State Drives, Intel® 40Gbe Networking, Intel® Advanced Vector Extensions 2 (Intel® AVX2),
and Intel® Transactional Synchronization Extensions (Intel® TSX). Click the links to see the detailed results! indicates new since May
2015:
Design rule checking and physical verification
• Up to 25% faster performance
TECHNICAL COMPUTING
Topological Data Analysis insight
• Up to 1.85X performance increase
New Mixed Analytics benchmark
BIG DATA ANALYTICS
CORE BUSINESS & TELECOMMUNICATIONS
OLTP and OLAP transactions workload
HEALTH & LIFE SCIENCES
DNA sequencing pipelines
• Up to 66% faster performance
FINANCIAL SERVICES
Risk analytics calculations
• Up to 2.48X faster performance
ENTERPRISE DATABASE
Online transaction processing
• Up to 6X more transactions per second.
+ =
Powerful
software used
to gain
meaningful
insights
Compelling
reasons to
optimize
applications
6
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components,
software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the
performance of that product when combined with other products. For more complete information visit http://www.intel.com/performance.
N E W

 37 Proof Points, spanning a range of key business vertical, 7 of which reflect the benefits of using
other Intel components beyond the processor, known as Better Together proof points.
 7 2x2s highlighting how new hardware and new, optimized software generate significant
performance gains.
 Links to marketing assets:
 Videos
• Software AG/Apama* – High-Speed, Streaming Analytics from Software AG
• SAS* – Innovating the Data Frontier
• Vital Images* – Advanced Medical Imaging Speeds Insights
 Infographics
• AYASDI – Powerful Insights from Complex Data Infographic
• Microsoft SQL Server 2014* – Business Cases for Upgrading to Intel® Xeon® Processor E7 Family
• Software AG/Apama – Improving Real-Time Event Stream Data Analysis
• SAP HANA* – Uptime, Crunch Time, Real Time
 Solution Briefs
• Microsoft SQL Server* 2014 – Microsoft SQL Server* 2014 and Intel® Xeon® Processor E7 v3 Family
• Nomura Research Institute – NRI BOS and Intel Deliver Real-Time Business Insight
• Epic*/InterSystems*/VMware* – InterSystems, VMware Increase Database Scalability for Epic EMR*
• IBM DB2* – Capture & Capitalize on Business Intelligence
• SAP HANA - Ask for More From Your Data
• SAS – Better Performance & Scale For New Analytics
• Software AG/Apama – Apama Streaming Analytics* with the Intel® Xeon® Processor E7 v3
For public presentations and more information, visit www.intel.com/XeonE7SoftwareSolutions
What’sInside?
7

PlatformValueofthe
Intel®Xeon®processorE7-8800/4800v3
Accelerate Time To Insights With The Latest Intel Hardware
8

*Other names and brands may be claimed as the property of others
Technical Computing
4-Socket
Business Processing
4-Socket
21 New World Records1
Infrastructure Virtualization
4 & 8-Socket
Decision Support
4-Socket
General Computing
2, 4 & 8-Socket
Decision Support
4-Socket
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions.
Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For
more complete information, visit http://www.intel.com/performance/datacenter. Intel does not control or audit third-party benchmark data or the web sites referenced in this document. You should visit the referenced web site and confirm whether referenced data are accurate.
1. Twenty-one performance world records as of May 5, 2015.
For more information about performance and benchmark results, visit www.intel.com/E7v3records.

Intel®Xeon®ProcessorE7v3FamilyHighlights
Up to 20% more cores1
Up to 20% more cache1
Up to 70% more query
session per hour for
better business insight2
Performance
(vs. previous generation)
Intel® TSX
Intel® AVX 2.0
Cache Monitoring
Intel® VMCS shadowing
New Features
Enhanced MCA Recovery Gen 2
Address Range Memory
Mirroring
Multiple Rank Sparing
DDR4 Recovery for Command
& Address Parity Errors
New Reliability
Features
DDR4 and DDR3 supported
to provide headroom for
future growth as your
workloads grow
Largest Memory
Footprint
1 - Intel® Xeon® processor E7 v3 family compared to previous gen Intel® Xeon® processor E7 v2 family.
2 - Configuration details: SAS Mixed Analytics workload measuring sessions per hour using SAS Business Analytics* 9.4 M2 on Red Hat* Enterprise LINUX 7. Baseline 1.0: 4S Intel Xeon processor E7-4890 v2, 512 GB DDR3-1066 memory, 16x 800 GB Intel®
SSD DC S3700, scoring 0.11 sessions/hour. Up to 1.72x more sessions per hour: 4S Intel Xeon processor E7-8890 v3, 512 GB DDR4-1600 memory, 4x 2.0 TB Intel® SSD DC P3700 SSDs + 8x 800 GB Intel® DC S3700 SSDs, scoring 0.19 sessions/hour.
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software,
operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that
product when combined with other products. For more complete information visit http://www.intel.com/performance
10

11
KeyGenerationalPerformanceIncreases
Intel®Xeon®ProcessorE7v3Family
Up to 39% more OLTP database
transactions/time1
Up to 6x more
in-memory OLTP transactions w/
Intel® TSX optimizations2
Up to 70% additional analytical
sessions/hour3
Up to 39% application
throughput gain for virtualized
machine (VM) infrastructure
consolidation4
Business Processing Business Analytics Virtualization
1,2,3,4 – For configuration details, go here. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using
specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your
contemplated purchases, including the performance of that product when combined with other products. For more complete information visit http://www.intel.com/performance.

12
BusinessOpportunities
Accelerating Insight with Real-Time Analytics
Delivering High Reliability and Uptime
Better Total Cost of Ownership
Achieving Operational Excellence
Enhanced Performance in the Data Center

Softwareplatformvalueproposition
With software solutions running on the Intel® Xeon® processor E7 v3 family, you can:
 Gain actionable business insights for quick decision making and competitive advantage via large
scale, real time in-memory analytics capabilities.
 Realize up to a 1.7X1 generational gain for analytics workloads and up to 6X2 more transactions
per minute with optimized software and the new Intel TSX enabled.
 Capitalize on operational efficiency from a 9:1 consolidation and 12X performance
improvements over 5-year old systems, with smaller datacenter footprint and lower TCO3.
 Enjoy world class reliability and 99.999% uptime for your most critical workloads, with Intel® Run
Sure Technology and new RAS features exclusive to the Intel® Xeon® processor E7 v3 family4.
 Ensure headroom for growth as your workloads expand, with support for both DDR4 and DDR 3
memory.
13
6X
UP TO
PERFORMANCE
INCREASE WITH
NEW SOFTWARE
AND INTEL® TSX
ENABLED
1,2,3 – For configuration details, go here. 4 - Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration. No computer
system can be absolutely secure. Check with your system manufacturer or retailer or learn more at www.intel.com.
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests are measured using specific computer systems, components, software, operations and functions. Any change to
any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. Visit
http://www.intel.com/performance/datacenter for more performance details.
Tame big data. Unleash amazing business possibilities.

Softwareplatformvalueproposition
14
S.
What does the Intel® Xeon® processor E7 v3 family mean
for your business?
 Analyze complex global sales data in minutes not hours.
 Personalize the retail shopping experience in real - capitalize on
opportunities to up-sell and cross-sell.
 Analyze ticks, tweets, satellite imagery and weather trends to
adjust energy and utility operations, financial trading, or local
public services in real time.
 Turn clinical data from EHR and other systems to improve
treatment plans and patient outcomes.
 Deliver real-time, business-critical services such as healthcare,
energy, financial trading, and logistics applications, by scaling
memory to optimize resources to workload.
Turn real-time insights into real business advantages.
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components,
software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the
performance of that product when combined with other products. For more complete information visit http://www.intel.com/performance.

OperatingSystemOptimizations
15

Platform
• Platform and KVM support for all Intel® Xeon® processors.
• Expanded support for PCI*-AER, Power Management, EVB and DCB,
Intel® AVX2 and USB 3.0.
Security
• Trusted boot that allows for Intel® Trusted Execution Technology
(Intel® TXT).
• Expanded secure platform support including attestation service stack.
Performance and Scalability
• Supports up to 4,096 CPUs and 64TB of RAM as a Host.
• Supports up to 64 cores and up to 256GB of RAM as a Guest.
• KVM supports the highest reported number of x86 guests and scales
up to 160 vCPUs, 2TB of RAM and 1024 disks.
Storage
• Advanced Host Controller Interface (AHCI) drivers.
• New scalable storage architecture: virtio-scsi.
• Dedicated threads for direct I/O: Data-plane mode in virto-blk.
• Intel® Rapid Storage Technology using open source MD RAID package
• /mdadm utility.
RAS
• All Intel® Xeon® RAS features.
• Live snapshots (available with Red Hat* Enterprise Linux OpenStack
Platform).
• Live storage migration and delete (available with Red Hat Enterprise
Linux OpenStack Platform).
Networking
• Open vSwitch* (available with Red Hat Enterprise Linux OpenStack
Platform).
• Edge Virtual Bridging (EVB).
• Data Center Bridging (DCB).
• SR-IOV support for VF enabled networking cards.
RedHat*EnterpriseLinux7
OptimizedfortheIntel®Xeon®E7v3Family
16
Information provided by the vendor has not been vetted and this does not constitute an endorsement.

SUSE Linux Enterprise* 12
PLATFORM
 CPU: SMEP, Intel® AVX2, Xsave/Xrstor, Xsaveopt, Intel® Turbo 2.0,
package thermal, power limit notification, energy performance
bias.
 UEFI: Secure Boot
 Networking: IPv6
 Storage/IO: MDRAID 0, 1, 5, 10, NVMe, PCIe* 3.0, USB 3.0
UPTIME
 RAS: Corrected error reporting (CMCI), advanced mcelog, MCA
Recovery, HWPoison, PCI AER, AER recovery interface with
APEI/GHES
 Full system rollback: btrfs and snapper integrated with
bootloader.
 HA and geo-clustering with SLE extensions.
 Kernel live patching with SUSE Linux Enterprise Live Patching*.
SUSE Linux Enterprise* 12 is an advanced foundation for
deploying highly available enterprise-class IT services in mixed
environments with improved uptime, better efficiency, and
accelerated innovation.
SUSE Linux Enterprise
 UNIX to Linux Migration
 Mission-Critical Computing
 Integrated Systems
 Big Data
 Virtualization and Cloud
SUSELinuxEnterprise*
OptimizedfortheIntel® Xeon® processorE7v3family
SCALABILITY AND PERFORMANCE
 Supports up to 8,192 CPU thread, 64 TiB RAM SSE 4.2, I/OAT3,
NUMA, UEFI 2.3.1, Turbo AES-NI
 Always running Local APIC timer 15-core support, 46/48-bit memory
addressability
 VMCS Shadowing
EFFICIENCY
 Flexible Virtualization : APIC-V, Flex Migration, real mode EPT (Xen
and KVM), Pause Loop Exiting (Xen and KVM), Native VT-d2, TXT
(Xen)
 Energy Efficiency: Multicore Power Management, DBS, Deep C-states
Reference: Refer to https://www.suse.com/releasenotes/x86_64/SUSE-
SLES/12/ for a complete description.
17

 Support for Intel® Virtualization Technology
 Ultimate Performance
 Network and block device hot plugging
 Virtual machine management and orchestration tools
 Scalability, live migration
 High Availability
 Full support for both leading open source hypervisors, KVM and Xen*
 Support for Linux containers (libvirt-lxc) and Docker
 Multiplatform support
 SUSE Linux Enterprise Server* 10, 11 and 12
 Novell Open Enterprise Server* 11
 NetWare* 6.5
 Windows Server* 2003, 2008, 2008 R2, 2012, 2012 R2
 Windows Vista*, 7, 8, 8.1
 Red Hat* Enterprise Linux 5, 6, 7
 Host Limits with SLES 12:
 8192 logical CPUs
 16 TiB RAM
 Guest Limits with SLES 12:
 256 CPU threads
 4 TiB RAM
SUSELinuxEnterprise*
KeyVirtualizationFeatures
SUSE Linux Enterprise
 UNIX to Linux Migration
 Mission-Critical Computing
 Integrated Systems
 Big Data
 Virtualization and Cloud
18

VMwarevSphere*6.0
OptimizedfortheIntel® Xeon® E7v3Family
VMware ESXi* 6.0
• Optimized to take advantage of Intel® Xeon® processor performance and efficiencies.
Supports new Intel® Xeon® processor E7 v3 family features:
• Intel® Advanced Encryption Standard Instructions (Intel® AES-NI) to accelerate
encryption/decryption.
• Intel® AVX2.0 to accelerate PCoIP for DVI market.
• A/D bits in EPT to support MP Record/Replay.
• VMFUNC to support nested virtualization (HW assisted nested virtualization –
VMware has been implementing nested virtualization with SW).
• MCA Core Error Recovery to provide better fault tolerance.
Continues to support previous platform features:
• 320 vCPU
• 4TB DRAM support
• OS Guard and Secure Key
• Hot-plug PCIe*
• Intel® AES-NI

BigDataAnalytics
21
Core Business
& Telecomm
Enterprise
Database
Financial
Services
Technical
Computing
Health
& Life Sciences
Up to
Faster1
85%
POWERFUL NEW PLATFORM FOR BIG
DATA INNOVATION
Leading software vendors are redefining what’s
possible – and what’s practical – in data-driven
industries and research disciplines. From
massively scalable real-time bidding to predictive
analytics and deep, highly-visual exploration of
complex data sets, vendors are delivering
breakthrough capabilities. The Intel® Xeon®
processor E7 v3 family and the Intel® Solid-State
Drive Data Center Family are playing a key role,
by taking speed, scalability, and value to new
heights. See the following slides to learn more
and to see more impressive results.
1 – Demonstrated with Ayasdi* proof point.

0
1
4S Intel® Xeon® processor E7-4890 v2
4S Intel® Xeon® processor E7-8890 v3 + Intel AVX2
AYASDI Machine Intelligence increased throughput
from 3.67 to 6.8 analyses/hour, 1.85X, with 4S Intel®
Xeon® processor E7-8890 v3, Intel® AVX2, and Intel®
SSD P3700
AYASDI*
Machine Intelligence* for the Enterprise
“Moving to Intel® Xeon® processor E7-8890 v3 processors and fully
leveraging Intel® Advanced Vector Extensions 2 (Intel® AVX2) and Intel®
Solid-State Drives (Intel® SSDs) improved the performance of our Machine
Intelligence* software by up to 85%!”1
Lawrence Spracklen - VP of Engineering at Ayasdi
22
www.ayasdi.com
 Ayasdi’s Machine Intelligence* software employs Topological Data Analysis
(TDA), to simplify the extraction of knowledge from even the most complex
data sets confronting organizations today – converting data into business
impact.
 Ayasdi’s Fortune 1000 clients use Machine Intelligence to find patterns and
relationships in complex data that elude other approaches. From risk
management to segmentation to anomaly detection and model validation,
clients across industries are extracting new insights from their rapidly
expanding data assets.
Big Data Analytics
Up to
85%
faster!!
1 - Testing conducted on AYASDI Machine Intelligence* software comparing 4S Intel® Xeon® Processor E7-8890 v3, 2.5GHz, DDR4 and Intel® AVX2 with 4S Intel® Xeon® Processor E7-4890 v2, 2.8GHz, DDR3. Baseline
configuration details: RHEL* 7, Intel® SSD P3700. Testing done by AYASDI. For complete testing configuration details, see SLIDE 85. Software and workloads used in performance tests may have been optimized for performance
only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may
cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other
products. For more complete information visit http://www.intel.com/performance.
Intel Better Together: XEON + AVX2 + SSD
Intel® Xeon® Processor E7-8800/4800 v3 Family
Powerful insights from complex data using Machine
Intelligence
Workload: Emulates appliance behavior under load, scaled down to
accommodate the small size of the test cluster size and limited
memory resources. Performance is measured in terms of 1) the
end-to-end latency of the analyses, and 2) the overall throughout of
analyses that can be supported by the cluster, with acceptable
latency.
Normalizedperformance

0
1
Intel® Xeon® processor E7-4890 v2 + Intel® SSD S3700
Intel® Xeon® processor E7-8890 v3 + Intel® SSD P3700/S3700
SAS*
Business Analytics*
“Mixed SAS Workloads can simultaneously generate extremely high and
prolonged CPU utilization, and heavy large-block sequential IO demand.
Testing with the Intel® Xeon® processor E7-8890 v3 and Intel® P3700 SSD
storage has shown significant performance improvement for the SAS Mixed
Analytic Workload performance test. Workload job run time was reduced,
while lowering CPU cycle utilization. The resultant combination of
computational and IO performance improvements utilizing the Intel® Xeon®
processors and Intel® SSD storage are impressive and will benefit SAS
customers.
Tony Brown - Principal Software Developer, SAS R&D
23
www.sas.com
 SAS* applications provide an integrated environment for predictive and
descriptive modeling, data mining, text analytics, forecasting, optimization,
simulation, experimental design and more.
 Customers can now run more complex analyses in less time and across
larger data sets to gain deeper insight into critical business issues.
Speed time to insight for better decision making!
Big Data Analytics
SAS Analytics* new mixed workload improvement
from .11 to .19 sessions/hour, 1.72X, with 4S Intel®
Xeon® Processor E7-8890 v3 and Intel® SSD DC
P3700 and S3700
Up to
72%
faster!!
1 - Testing conducted on SAS Business Analytics* 9.4 M2 jobs-per-hour comparing 4S Intel® Xeon® Processor E7-8890 v3 + Intel® SSD DC P3700/S3700 with 4S Intel® Xeon® Processor E7-4890 v2 + and Intel® SSD DC S3700.
Testing done by Intel. For complete testing configuration details, see SLIDE 85. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests,
such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other
information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit
http://www.intel.com/performance.
Intel Better Together: XEON + SSD
Workload: New SAS mixed analytic native benchmark. The SAS new
mixed analytic workload represents the type SAS customers use
today.

0
1
24
Software AG*
Apama Streaming Analytics*
“The Intel® Xeon® processor E7-8890 v3 improved event throughput for
Apama Steaming Analytics* by over 55% and location updates increased by
20% per second. These are fantastic results for us.”1
Dr. Mark Horsburgh - Senior Software Architect, Software AG Big Data Analytics
www.softwareag.com
1 - Testing conducted on Apama* software comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Software AG. For complete configuration details, see SLIDE 85.
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems,
components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated
purchases, including the performance of that product when combined with other products. For more complete information visit http://www.intel.com/performance.
Higher throughput helps customers build powerful,
flexible, cost effective analytics solutions.
Software AG Apama* increased from 43.9 to 70M
events per second, 1.59X, with 4S Intel® Xeon®
processor E7-8890 v3
Up to
59%
faster!!
Workload: The workload is a micro-operation that is performed very
often inside Apama. The workload receives events in the pipeline
stage and publishes an output from the Apama correlator.
 Apama Streaming Analytics is a complex event processing engine. It
filters, analyzes, and acts on data.
 Increased event throughput helps improve scale, solution flexibility, and
customer confidence.
 Customer use cases include:
 eCommerce
 Location aware customer experience management
 Complex event processing

25
CoreBusiness&Telecommunications

CoreBusiness&telecommunications
26
Big Data
Analytics
Enterprise
Database
Financial
Services
Technical
Computing
Health
& Life Sciences
Up to
Faster1
106%
SCALE AND ADAPT YOUR MOST CRITICAL
APPLICATIONS
Change is never-ending in IT, but one thing
remains constant: Core business and
telecommunications applications must
continuously scale and adapt. The Intel® Xeon®
processor E7 v3 family delivers functionality to
more users with improved performance, tighter
security, and improved cost models. The following
slides show how much additional performance
you can expect for some of today’s most critical
applications.
1 – Demonstrated with NRI BOS* proof point.

0
1
2
4S Intel® Xeon® processor E7-4890 v2 + Intel® AVX
4S Intel® Xeon® processor E7-8890 v3 + Intel® AVX2
27
Nomura Research Institute, Ltd.*
Business Oriented Solution (BOS)
“NRI* is excited that the Intel® Xeon® processor E7-8890 v3 improved BOS’s
NUMA optimized application performance. More CPU cores and Intel® AVX2
really boosted performance to amazing levels compared to the previous
generation processors”
Yuzo Ishida – Senior Application Engineer, NRI (Architect of BOS)
Core Business & Telecommunications
 BOS* is the new generation Big Data platform, optimized to utilize
multicores and Intel AVX2, to maximize performance on the Intel Xeon
processor E7-8890 v3.
 BOS* handles OLTP and OLAP transactions on a single database scheme
without any data mart or pre-manipulated redundant data source.
 Increased memory bandwidth in the Intel Xeon processor E7-8890 v3
allows better utilization of computational resources, and significantly
improved runtimes when data increased.
 Real-time enterprise architecture that scales with NUMA optimizations.
www.nri.com
Data-oriented business insights delivered twice as fast speeds
up decision-making and lowers TCO!
1 - Testing conducted on Nomura Research Institute, Ltd.* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by NRl. For complete testing configuration details, see SLIDE 85.
Up to
106%
faster!!!
Workload: Real-time stock calculation; Benchmark schema derived
from TPC-H* queries.
NRI BOS* improved OLAP transaction processing time
from 62 to 30 milliseconds, 2.06X, with 4S Intel®
Xeon® processor E7-8890 v3 and Intel® AVX21
Intel Better Together: XEON + AVX2
Normalized
performance

0
1
28
www.ibm.com
 The 4S Intel® Xeon® processor E7-8890 v3 enabled IBM InfoSphere
Information Server DataStage* v11.3.1 to realize up to 40% improvement
in throughput, compared to the previous processor generation.1
 With the ability to move data up to 40% faster between repositories,
customers can keep up with the explosive growth in data without
experiencing significant bottlenecks in data transfers.
IBM*
IBM InfoSphere Information Server DataStage* v11.3.1
IBM InfoSphere Information Server DataStage* v11.3.1 is a market-leading
data integration platform that helps you understand, cleanse, transform and
deliver trusted information to your critical business initiatives, such as big
data, master data management and point-of-impact analytics.
Move data up to 40% faster between repositories – minimize
bottlenecks from ongoing data growth!
1 - Testing conducted on IBM InfoSphere DataStage 11.3.1* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by IBM. For complete testing configuration details, see SLIDE
85. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems,
IBM InfoSphere* increased throughput
performance from 601 to 842.6 MB/sec, 1.4X,
with 4S Intel® Xeon® processor E7-8890 v3
Up to
40%
faster!!
Workload: Typical Data Warehousing Integration Scenario; based on
Brokerage Firm Business Scenario that performs extract, transform
and load operations (ETL). CPU intensive ETL test (DimTrade) used.
Data volume size is 53GB.

0
1
29
www.ibm.com
 Multi-core and vector optimized algorithms paired with massive parallelism
exploit the high core count of Intel® Xeon® processors with SIMD
technology.
 IBM Informix* Warehouse Accelerator (IWA) can speed up even the most
complex warehouse queries, without changes to business applications.
 The 4S Intel® Xeon® processor E7-8890 v3 enables IWA to further shorten
query times, and deliver results faster, by up to 28%.
IBM*
IBM Informix*
IBM Informix* Warehouse Accelerator loads data completely into system
memory in a compressed form, using a special columnar scheme developed
by IBM. As queries come in, the Informix database passes them to the
Informix Warehouse Accelerator which can scan billions of rows of data in
seconds to return immediate results.
1 - Testing conducted on IBM Informix* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by IBM. For complete testing configuration details, see SLIDE 85. Software
and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems,
components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your
Up to
28%
faster!!
Workload: Analytical queries on a large set of sensor time series
data, which originates from publicly available U.S. government data
(National Water Information System).
IBM Informix* increased the geomean performance
from 10628.58 to 13624.74 selects/day, 1.28X, with
Get your queries answered up to 28% faster!

0
1
30
Neusoft*
SaCa Aclome*
“The Intel® Xeon® processor E7-8890 v3 provides a great 1.21X performance
boost compared to the previous generation. It supports our product to deliver
better user experiences to our customers and it definitely will be one of our
preferred platforms.”
Ji Yong – Director of Platform Software Division, Neusoft
 Neusoft SaCa Aclome* is a visualized, automatic and agile cloud
management environment enabling customers to deploy, monitor and
manage cloud applications in a flexible, convenient manner.
 The Intel® Xeon® processor E7-8890 v3 computing capability helps
Neusoft SaCa Aclome to deliver more services in less time.
 For Neusoft SaCa Aclome workload, customer can get up to 1.21x
performance gain on Intel® Xeon® processor E7-8890 v3 (2.5G)
compared to the Intel® Xeon® processor E7-4890 v2 (2.8GHz).
www.neusoft.com
Neusoft SaCa Aclome* increased performance
from 2,794 to 3,381 transactions per second,
1.21X, with 4S Intel® Xeon® processor E7-8890 v3
1 - Testing conducted on Nuesoft SaCa Aclome* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Nuesoft. For complete testing configuration details, see SLIDE 85.
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer
systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your
Workload: Neusoft load test tool to emulate report generation
throughput and response time, a heavy workload request.
Up to
21%
faster!
Deliver more services in less time, with faster cloud building,
computing, storage and network resources!
Normalized
performance

0
1
31
TYDIC*
OCS*
“TYDIC* customers rely on OCS* to support critical business functions and
have strict performance and reliability requirements. The 1.21X performance
gains provided by the Intel® Xeon® processor E7-8890 v3 will help our
customers achieve higher value at lower cost. This is one of the ideal
deployment platforms for OCS.”
Liu Yu – Director of Billing product, TYDIC
 TYDIC* is a leading ISV for Telecom BOSS in China. OCS2.0 is its online
charge system based for Telecom Carriers.
 The Intel Xeon processor E7-8890 v3 computing capability helps
TYDIC OCS to support more customer transactions per second.
 The up to 1.21X performance gain enables telecom providers to service
more customers more quickly to better support tailoring offers to
customers, delivering loyalty programs, and for more efficient billing
management.
www.tydic.com/
TYDIC OCS* increased performance from 411K
to 501K transactions per second, 1.21X, with 4S
Intel® Xeon® processor E7-8890 v3
Up to
21%
faster!
1 - Testing conducted on TYDIC OCS* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by TYDIC. For complete testing configuration details, see SLIDE 86. Software
Workload: OCS workload uses stressing tool to simulate typical
business operations in OCS system.
Service more customers faster, achieving greater value at a lower
cost!

0
1
32
Yonyou*
ERP-NC*
“Based on the Intel® Xeon® processor E7-8890 v3, the ERP-NC* application
showed an impressive performance boost compared with the previous
processor generation. The great performance and reliability fully satisfy NC
requirements for its mid or large-scale enterprise customers. It is the ideal
deployment platform for NC.”
Zhao Yashun – GM of Industry Chain Business Division, Yonyou
 Yonyou* is the leading ERP software supplier in Asia-Pacific and China.
ERP-NC* is a J2EE-based ERP for large-scale companies.
 The Intel Xeon processor E7-8890 v3 improves Yonyou ERP-NC
customer transactions per second by up to 21%.
 For the Yonyou NC workload, customers can experience an up to 1.21X
performance gain with the Intel® Xeon® processor E7-8890 v3 (2.5GHz)
compared to the Intel® Xeon® processor E7-4890 v2 (2.8GHz).1
http://www.yonyou.com/
Yonyou ERP-NC* increased performance from
1,567 to 1,896 transactions per second, 1.21X,
1 - Testing conducted on Yonyou ERP-NC* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Intel. For complete testing configuration details, see SLIDE 86.
Up to
21%
faster!
Over 300 additional customer transactions per second!
Workload: NC stress tool simulates typical business operations in
NC; pressure focuses on application server side.
.

0
1
33
Yonyou*
U9*
“Based on the Intel® Xeon® Processor E7-8890 v3, Yonyou U9 showed a great
performance boost compared with previous generation processor. This will
give our U9 customers a better user experience and it is one of our preferred
platforms.”1
Zhao Yashun - GM of OEM & Industrial Chain Division, Yonyou
 Yonyou is the leading ERP software supplier in Asia-Pacific and China.
UFIDA U9 is its .Net-based ERP for large-scale company and group.
Yonyou U9 to support more transactions per second for its customers.
www.yonyou.com/
Copyright © 2015 Intel Corporation. All rights reserved. *Other names and brands may be claimed as the property of others.
1 - Testing conducted on Yonyou U9* comparing 4S Intel® Xeon® Processor E7-8890 v3 to 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Yonyou and Intel. For complete configuration details, see SLIDE 86.
Yonyou U9* increased transactions per second from
103 to 125, up to 1.21X faster, with the 4S Intel®
Xeon® processor E7-8890 v3
Workload: The U9 workload uses a stress tool to simulate typical
business operations in U9 and the pressure focuses on the
application server side. CPU utilization is around 85%.
More transactions per second for the leading ERP
software supplier in Asia-Pacific and China!
Up to 21%
Faster!
N E W

0
1
34
TongTech*
TongWeb*
“With the Intel® Xeon® Processor E7-8890 v3, TongTech TongWeb showed up
to a 1.21X performance boost compared with previous generation processor.
This will definitely improve the TongWeb user experience.”1
Yang Jiawei - TongTech TongWeb Marketing Director Core Business & Telecommunications
 TongTech is the leading middleware vendor in PRC and its products
include several middleware platforms and application packages.
TongWeb is a leading localized J2EE Application Server* in PRC, and it
provides highly-available, scalable and manageable platforms for J2EE-
based enterprise applications.
TongTech TongWeb to finish more requests in less time for its
customers.
www.tongtech.com
1 - Testing conducted on TongTech TongWeb* comparing 4S Intel® Xeon® Processor E7-8890 v3 to 4S Intel® Xeon® Processor E7-4890 v2. Testing done by TongTech and Intel. For complete configuration details, see SLIDE 86.
TongTech TongWeb* increased Enterprise jAppSever
Operations Per Second (EjOPS) from 20,913 to
25,397, up to 1.21X faster, with the 4S Intel® Xeon®
Workload: SPECjEnterprise* 2010, a full system Java EE* 5.0 servers
and supporting infrastructure performance measurement and
characterization benchmark.
TongTech TongWeb delivers thousands more EjOPS,
improving the user experience!
Up to 21%
Faster!
N E W

0
1
35
TongTech*
TongIntegrator*
“Taking advantage of the Intel® Xeon® Processor E7-8890 v3, we see a great
performance boost on TongTech TongIntegrator compared with the previous
generation processor. We will surely recommend this platform to our
TongIntegrator customers.”1
Yang Jiawei – Tongtech Marketing Director
 TongTech is the leading middleware vendor in PRC, and its products
include several middleware platforms and application packages.
TongIntegrator is a J2EE*-based middleware providing data conversion
and integration capability among different IT systems.
 The 4S Intel® Xeon® processor E7-8890 v3 computing capability helps
Tongtech TongIntegrator to support more transactions per second for its
customers.
www.tongtech.com
1 - Testing conducted on TongTech TongIntegrator* comparing 4S Intel® Xeon® Processor E7-8890 v3 to 4S Intel® Xeon® Processor E7-4890 v2. Testing done by TongTech and Intel. For complete configuration details, see SLIDE 86.
TongTech Tongintegrator* increased transactions per
second from 95.9 to 116, up to 1.2X faster, with the
Workload: TongIntegrator workload simulates typical data
integration scenarios, i.e., fetch data from source db, conversion
customized by end-user, store converted data into destination
database.
TongTech TongIntegrator can now deliver more services to
customers in less time!
Up to 20%
Faster!
N E W

0
1
36
Digital China*
Sm@rtESB*
“The Intel® Xeon® Processor E7-8890 v3 demonstrates its technology
advantage, great performance and reliability. It helps Digital China
Sm@rtESB* to gain 1.19X performance improvement compared to the
previous processor generation, and will provide better user experiences to
our customers.”
Yu Xiang – Senior Architect, Digital China
 Digital China Sm@rtESB* is an enterprise IT infrastructure for Service-
Oriented Architecture (SOA).
 The Intel Xeon processor E7-8890 v3 computing capability helps
Digital China Sm@rtESB to support more users while providing
services faster.
 For Digital China Sm@rtESB workload, customer can get 1.19X
performance gain on Intel Xeon processor E7-8890 v3 (2.5G) compared
to the Intel Xeon processor E7-4890 v2 (2.8GHz)
www.digitalchina.com
Digital China Sm@rt ESB* increased performance
from 26.8K to 31.9K transactions per second,
*Other names and brands may be claimed as the property of others.
1 - Testing conducted on Digital China Sm@rtESB* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Digital China. For complete testing configuration details, see SLIDE 86.
Up to
19%
faster!
Workload: Digital China stressing tool to simulate processing
messages with industrial standard format among different IT system
destinations.
Greater business efficiency and better user experiences!

0
1
37
Huawei*
FusionSphere*
“Based on the Intel® Xeon® Processor E7-8890 v3, Huawei FusionSphere*
showed an impressive performance boost compared to the previous
processor. The great performance, reliability and stability fully satisfy our
customers’ strict requirements. It is the ideal deployment platform for Huawei
FusionSphere.”
Chen Kun - Director of Huawei Cloud Computing & Virtualization Product
Huawei customers’ requirements satisfied more reliably and faster!
1 - Testing conducted on Huawei FusionSphere* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Intel. For complete testing configuration details, see SLIDE 86.
www.huawei.com
Up to
18%
faster!
Workload: Stressing tool to simulate typical enterprise solution
usage scenario.
Huawei FusionSphere* increased performance
from 47,924 to 56,702 operations per second,
 Huawei FusionSphere* is a backend virtualized cloud solution.
FusionSphere uses virtualization technology to virtualize the backend
server resources and provides agile and efficient HW infrastructure
environment for enterprise solutions.
 Intel® Virtualization Technology on the Intel® Xeon® processor E7-8890
v3 helps Huawei FusionSphere to gain high performance and high
reliability in a virtualization environment.
Normalized
performance

 NARI Technology Development Co., Ltd.* is a leading provider of power and automation
technologies and the major solution provider for the State Grid in China. Its real-time
database is used to collect and store data from thousands of sensors across the state
grid and is the interface for querying and analyzing the data. With the 4S Intel® Xeon®
Processor E7-8890 v3, NARI increased transactions per second from 1,863 to 2,236, up
to 1.2X faster, compared to the previous generation processor1. www.naritech.cn/en/
38
Out-of-the-Box Performance Increases
1,2, - Testing conducted on ISV applications comparing 4S Intel® Xeon® Processor E7-8890 v3 to 4S Intel® Xeon® Processor E7-4890 v2. Testing done by ISVs and Intel. For complete configuration details, see SLIDE 86. Software
 Neusoft HIS* (Hospital Information System) is a key healthcare solution from Neusoft. It includes
information systems for outpatient, inpatient, clinic, medical techniques, medicines, logistics, general
management and statistics and analysis. With the 4S Intel® Xeon® Processor E7-8890 v3, Neusoft HIS
increased transactions per second from 173 to 206, up to 1.19X faster, compared to the previous
generation processor2. www.neusoft.com
Leading Solutions Providers in China rely on Intel® Xeon® Processors to Deliver Results!
N E W

EnterpriseDatabase
40
Big Data
Analytics
Core Business
Applications
Financial
Services
Technical
Computing
Health
& Life Sciences
Up to
Faster1
6X
THE RIGHT FOUNDATION FOR A
SOFTWARE-DEFINED INFRASTRUCTURE
The Intel® Xeon® processor E7 v3 family is
delivering powerful new capability and value for
scaling, automating, and securing data center
infrastructure. For example, SAP HANA* is
taking advantage of the enhancements for
online transaction processing (OLTP), provided
by Intel® Transactional Synchronization
Extensions (Intel® TSX), to deliver significantly
more transactions per unit of time. See the
following slides to learn more and to see more
impressive results.
1 – Demonstrated with SAP HANA* proof point.

0
1
2
3
4
5
6
4S Intel® Xeon® processor E7-4890 v2 (SPS08)
4S Intel® Xeon® processor E7-8890 v3 + Intel® TSX
SAP*
HANA* SPS09
“SAP HANA* delivers next-generation in-memory computing through an on-
going engineering collaboration between SAP and Intel to provide optimized
performance and reliability on Intel® Architecture, including the Intel® Xeon®
processor E7 v3 family. The latest SAP HANA SPS09 release implements a
number of leadership enhancements for online transaction processing
(OLTP) accelerated by Intel® Transactional Synchronization Extensions
(Intel® TSX). Intel TSX helped us to solve concurrency challenges for stress
workloads leading up to 6X total improvement in processing throughput.”
Daniel Schneiss - Head of SAP HANA Platform Development,
Executive Vice President, SAP SE
41
 Analyze and act upon business data almost as fast as it is being generated!
 Intel TSX helps improve performance of in-memory database processing,
delivering transactions over twice as fast as previous generation
processors, when combined with SAP HANA SPS09.
Enterprise Database
www.sap.com
1 - Testing conducted on SAP HANA SPS09* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Intel. For complete testing configuration details, see SLIDE 86.
Up to 6X
faster!!!!
Workload: SAP internal workload stressing HANA OLTP engine.
SAP HANA* increased from 14,327 w/SP8 to 84,721
transactions per minute, 6X, w/SPS09 and with 4S
Intel® Xeon® processor E7-8890 v3 and Intel® TSX
Intel Better Together: XEON + TSX + SW
Turn real time analytics into real business advantage with the
latest from Intel and SAP HANA!

0
1
4S Xeon® E7-4890 v2 + Intel® SSD DC 3700
4S Xeon® E7-8890 v3 + Intel® TSX + Intel® SSD DC 3700
Intel Better Together: XEON + TSX + SSD
42
SAP*
Adaptive Server Enterprise (ASE)
 Exploit the value of massive amounts of data at the speed of business
with SAP ASE* running on the latest Intel® Xeon® processor E7 v3 family.
 Proven reliability and security for mission-critical application
environments and high-performance transaction-based applications
involving massive volumes of data at a lower TCO.
 Up to 1.65X faster transactions with the 4S Intel® Xeon® processor E7-
8890 v3 with Intel® TSX enabled, and with the Intel® SSD DC P3700 vs.
the 4S Intel® Xeon® processor E7-4890 v2 and the Intel® SSD DC S37001.
Enterprise Database
Up to 65%
Faster!
1 - Testing conducted on SAP ASE* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by SAP. For complete configuration details, see SLIDE 87. Software and workloads used in
performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and
functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that
product when combined with other products. For more complete information visit http://www.intel.com/performance.
SAP ASE* increased throughput from 673,674 to
1,114,220, up to 1.65X, with the 4S Intel® Xeon®
Intel® TSX helps SAP ASE process more transactions
faster and at a lower cost.
Workload: TPC-C workload
www.sap.com/
N E W

0
1
43
Oracle*
MySQL*
“The new Intel® Xeon® processor E7-8890 v3 gives a visible extra
performance gain for MySQL* 5.7.7-rc running OLTP workload. We see close
to 70K TPS and up to 1.24X performance gain for a read-only scenario
compared to the previous processer generation.”
Tomas Ulin - VP Engineering, MySQL
 The Intel Xeon processor E7-8800/4800 v3, with enhanced
microarchitecture, additional enhanced cores and larger cache, delivers
a significant performance gain for database applications such as
MySQL*.
 More queries per second results in a faster response.
www.mysql.com
1 - Testing conducted on Oracle MySQL* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Intel. For complete configuration details, see SLIDE 87. Software and
workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems,
The world’s most popular open source database processes 10s
of thousands more queries per second!
Up to
24%
faster!!
Workload: Sysbench benchmark, OLTP test profile. Defined 100
tables with 10 million rows each, resulting in a database of 1 billion
entries and 260GB.
Oracle MySQL* increased throughput per second
from 54,988 to 68,663 queries, 1.24X, with the
Enterprise Database

44
Altibase*
HDB-DA*
“HDB-DA is highly-optimized for in-memory computing, providing low
latency to our customers. We see the Intel® Xeon® Processor E7-8890 v3
boosts HDB-DA performance and meet our customers’ in-memory needs.”
Joshua Choi – Head of Research and Development Dept., Altibase
 HDB-DA is an in-memory only database maximizing its performance by
Direct Attach Mode to eliminate overhead for optimal data processing
speed. For the uses of the most demanding speeds, in-memory only
DBMS is the way to go.
 The optimized and low-overhead Altibase HDB-DA in-memory
database utilizes the increased number of Intel Xeon Processor E7-
8890 v3 cores.
 HDB-DA achieves high-intensity transactional speeds due to the faster
Intel Xeon Processor E7-8890 v3 memory.
http://altibase.com
ISV logo
1 - Testing conducted on Altibase HDB-DA* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Altibase. For complete configuration details, see SLIDE 87.
0
1
Enterprise Database
Altibase HDB-DA* increased from 638,587 to
783,198 queries per second, 1.22X, with 4S Intel®
Xeon® Processor E7-8890 v3
Up to
22%
faster!!
Workload: Stock trade simulation. The performance metric is the
number of queries per second that reflects multiple aspects of the
capability of the system to process queries.
Faster speed and scalability for customer values

0
1
45
EXASOL*
EXASolution* 5.0
“The new Intel® Xeon® E7 -8890 v3 processor delivers even more
unprecedented analytic performance with no limits on your data. We see
up to 1.24X extra performance vs. the previous generation processor
running our latest EXASolution. It allows more users to access and analyze
big data simultaneously without compromising query performance.”1
Mathias Golombek – CTO, EXASOL
Enterprise Database
 The world’s fastest2, highly-parallel, scalable, in-memory clustered
database with column-based storage and compression. Ideally suited for
real-time big data reporting, analysis, and advanced analytics.
 EXASolution customers benefit from more concurrent users and lower
TCO of the 4S Intel® Xeon® E7 -8890 v3 processor compared to the
previous generation processor.
www.exasol.com
Up to 24%
Faster!
1 - Testing conducted on EXASolution* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Intel and EXASOL. For complete configuration details, see SLIDE 87. 2 - TCP-H
rankings Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems,
EXASOL EXASolution* increased queries per second
from 213 to 266, up to 1.24X, with the 4S Intel® Xeon®
Top per-node performance for the modern scalable
clustered in-memory database for analytics.
Workload: EXASolution TPC-H like, 100GB load
N E W

0
1
46
Baosight*
iHyperDB*
“With the rapid development of fine manufacturing technologies, our
solutions have to efficiently handle big data from frontend sensors, and this
required performance extends to our iHyperDB. It is amazing to see up to a
1.21X performance boost from the Intel® Xeon® Processor E7-8890 v3 for
Baosight iHyperDB, compared with the previous generation processor. It
provides a powerful and reliable infrastructure for our software solution.”1
Dong Wensheng – Baosight R&D Center GM
Enterprise Database
 Baosight is one of the leading manufacturer software solution vendors in
PRC. iHyperDB is a high-performance real-time database for steel,
petrochemical, coal and other manufactory customers.
 The 4S Intel® Xeon® processor E7-8890 v3 computing capability,
compared with the previous generation processor, helps Baosight
iHyperDB to support more transactions per second for its customers.
www.baosight.com/
1 - Testing conducted on Baosight iHyperDB* comparing 4S Intel® Xeon® Processor E7-8890 v3 to 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Baosight and Intel. For complete configuration details, see SLIDE 87.
Baosight iHyperDB* increased transactions per second
from 206K to 251K , up to 1.21X faster, with the 4S
Workload: The workload uses Baosight self-developed stressing
tool to simulate typical customer usage scenarios. Simulates typical
operations such as insert, query, update.
Baosight iHyperDB can now deliver tens of thousands
more transactions per second!
Up to 21%
Faster!
N E W

0
1
47
www.ibm.com
 BLU Acceleration* dramatically simplifies and speeds the delivery of
business insights from data, and takes full advantage of innovations in
microprocessor technology.
 Running IBM DB2 with BLU Acceleration on servers powered by the Intel
Xeon processor E7 v3 family provides an adaptive platform for mixed
workloads in multi-user environments.
 Intel® Xeon® processor E7-8890 v3 running IBM BLU* 10.5 FP5 increased
performance by up to 222X compared to Intel® Xeon® processor
E7-4890 v2 running IBM 10.1 FP12
IBM*
IBM DB2 with BLU Acceleration* v10.5
“IBM's Information Management unit and Intel have a long history of
innovating together. We continue to deliver improvements in performance,
reliability and security with DB2 on the Intel® Xeon E7-8890 v3 processor."
- Sean Poulley, IBM Vice President Product & Offering Management,
Analytics Platform
1,2 - Testing conducted on IBM DB2 10.5* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by IBM. For complete testing configuration details, see SLIDE 87.
Enterprise Database
Up to
21%
faster!!
Workload: DB2 v10.5, Proof of Performance and Scalability (POPS)
IBM DB2* improved query processing time from
63.79 to 52.39 seconds, 1.21X, with the with
Faster insights from more data for ultimate business agility!

48
MariaDB*
MariaDB 10.1*
“The new Intel® Xeon® processor E7 -8890 v3 continues the quad-socket
performance evolution. Powered by additional enhanced cores it improves
performance for traditional databases by up to 1.15x compared to the
previous generation. Our users and customers will get faster responses to
their complex queries, and can reduce TCO at the same time – so they can
realize even higher value with MariaDB”
Rasmus Johansson - VP Engineering, MariaDB Corporation
 MariaDB is a robust, scalable, and reliable SQL server.
 The Intel Xeon processor E7-8890 v3 with enhanced microarchitecture,
additional enhanced cores and larger cache give the performance boost
delivers a significant performance gain for traditional database
applications such as MariaDB.
 Faster response to complex queries, lower TCO.
https://mariadb.org
1 - Testing conducted on MariaDB 10.1* comparing Intel® Xeon® Processor E7-8890 v3 with Intel® Xeon® Processor E7-4890 v2. Testing done by Intel. For complete configuration details, see SLIDE 87. Software and
workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems,
0
1
Enterprise Database
Up to
15%
faster!
Workload: Typical R/O OLTP requests, 1M table size and 10M RW.
Multi-table scenario
MariaDB 10.1* increased throughput per second
from 49,323 to 57,001 queries, 1.15X, with the
with the 4S Intel® Xeon® processor E7-8890 v3
Faster response times and lower TCO on a popular, open-source
database!

 Kingbase ES* is a leading localized database in PRC, that is widely deployed in government, banking,
and education sectors to provide high-performance, high-availability data management for critical
applications. With the 4S Intel® Xeon® Processor E7-8890 v3, Kingbase ES increased transactions per
minute from 113K to 135K, up to 1.19X faster, compared to the previous generation processor2.
www.kingbase.com.cn
49
Out-of-the-Box Performance Increases
 AsiaInfo ADB* is a scalable OLTP database targeted for high-performance and mission critical business
operations such as online charge service (OCS) in the telecom industry. With the 4S Intel® Xeon® Processor
E7-8890 v3, AsiaInfo ADB increased transactions per second from 139K to 168K, up to 1.2X faster,
compared to the previous generation processor1. www.asiainfo.com
1,2,3 - Testing conducted on ISV applications comparing 4S Intel® Xeon® Processor E7-8890 v3 to 4S Intel® Xeon® Processor E7-4890 v2. Testing done by ISVs and Intel. For complete configuration details, see SLIDE 87/88.
Leading Solutions Providers in China rely on Intel® Xeon® Processors to Deliver Results!
 Kingdom Technology* is a leading ISV in the financial market industry in China, providing
transaction middleware kernels for applications. With the 4S Intel® Xeon® Processor E7-8890 v3,
Kingdom Technology STS* increased transactions per second from 11.8K to 14.2K, up to 1.2X
faster, compared to the previous generation processor3. www.szkingdom.com
N E W

FinancialServices
THE RIGHT PLATFORM FOR HIGH VOLUME
Volatile markets and growing regulatory demands
are generating new challenges across the financial
services industry. Software vendors such as
SunGard* are taking advantage of the Intel® Xeon®
processor E7 v3 family to help their customers
measure, manage and control risk more effectively.
The substantial performance gains provided by
these processors allow financial organizations to
make faster, smarter decisions in high-volume
trading environments.
In the coming weeks, more Financial Services
vendors will demonstrate how their applications
improve their performance with the Intel® Xeon®
processor E7 v3 family.
51
Big Data
Analytics
Core Business
& Telecomm
Enterprise
Database
Technical
Computing
Health
& Life Sciences
Up to
Faster1
46%
1 – Demonstrated with SunGard* proof point.

0
1
2
4S Intel® Xeon® processor E7-4890 v2 + SSD S3700
4S Intel® Xeon® processor E7-8890 v3 + SSD P3700
Up to 2.48X
Faster!!!
52
Kx Systems*
kdb+*
“The Intel® Xeon® E7-8890 v3 performance for the computational workload
provides a very significant increase (up to 2.48X) in performance over the
previous generation processor. These financial time-series calculations are
widely used and exploit the cores, registers and platform improvement of
the newest processor.”
Simon Garland – Chief Strategist, Kx Systems
Financial Services
 Kx users can significantly reduce the processing time, up to 2.48X, for
these typical financial workloads.
 The 4S Intel® Xeon® processor E7-8890 v3 increased memory
bandwidth allows better utilization of computational resources and
significantly improved runtimes.
 When combined with the Intel® SSD DC P3700, the 4S Intel® Xeon®
processor E7-8890 v3 delivers significant performance increases
across many financial workloads.
www.kx.com
www.STACresearch.com/kx
1 - Testing conducted on kdb+* comparing 4S Intel® Xeon® Processor E7-8890 v3 and Intel® DC SSD P3700 to 4S Intel® Xeon® Processor E7-4890 v2 and Intel® DC SSD P3700. Testing done by STAC*. 2,3 - *Kx and kdb+ are
registered trademarks of Kx Systems, Inc. STAC and all STAC names are trademarks or registered trademarks of the Securities Technology Analysis Center LLC.
For complete configuration details, see SLIDE 88. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and
MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and
performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit
http://www.intel.com/performance.
Kx Systems kdb+* decreased processing time from
13.69 to 5.5 seconds, up to 2.48X faster, with the 4S
Intel® Xeon® processor E7-8890 v3, Intel® AVX2, and
the Intel® SSD DC P37001
Workload: STAC-M3* 100-user intervalized statistics benchmark
(STAC-M3.ß1.100T.STATS-UI.TIME).
2 www.STACresearch.com/KDB150612
3 www.STACresearch.com/KDB140212
Intel Better Together: XEON + AVX2 + SSD
The Intel® Xeon® E7-8890 v3 processes more financial
calculations in less time!
2
3
N E W

0
1
53
SunGard*
Adaptiv* 13.1
“Through our strategic partnership with Intel, SunGard’s Adaptiv Analytics
platform has consistently taken full advantage of Intel’s processing
innovations and the Intel® Xeon® processor E7-8890 v3 is no exception. By
exploiting Intel® AVX2 instructions, we have yet again seen significant
increase in processing power. Our customers are not just demanding faster
results, they need to run more analysis to better understand risks. Their
appetite for more risk calculations is growing at a faster rate than their
hardware budgets so a platform that fully exploits all available processing
power is crucial to them.”
Chris Parkins – CTO, SunGard Adaptiv
Financial Services
 The increased throughput allows end users to perform more calculations
per second on a single machine resulting in a lower TCO.
 The Intel Xeon processor E7-8890 v3 increased core count provides
more resources for the workload to run on.
 Intel AVX2 instruction set and the two additional execution ports provide
increased performance.
http://financialsystems.sungard.com
Higher risk analytics throughput. More analyses processed in
less time!
1 - Testing conducted on SunGard Adaptiv 13.1* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Intel. For complete testing configuration details, see SLIDE 88.
Workload: Multiple Monte Carlo simulations each of 10K scenarios
on 10K long vectors of data.
SunGard Adaptiv* increased from 6.826M to 10.015M
calculations per second, 1.46X, with the 4S Intel®
Xeon® processor E7-8890 v3 and Intel® AVX21
Up to
46%
faster!!
Intel Better Together: XEON + AVX2

Health&LifeSciences
55
Technical
Computing
Manufacturing
Financial
Services
Up to
Faster1
66%
Big Data
Analytics
Core Business
& Telecomm
Enterprise
Database
TRANSFORMING HEALTHCARE
From better diagnostics to faster drug
discovery and deeper genomic analysis,
health care organizations and bio-medical
research teams are making major advances
in how we understand and treat disease.
See how the Intel® Xeon® processor E7 v3
family is helping leading software vendors
and research organizations achieve better,
faster results to improve medical outcomes
and accelerate the transition toward
personalized, data-driven health care.
1 – Demonstrated with Vital* proof point.

0
1
56
Vital Images, Inc.
VitreaAdvanced*
“The new Intel® Xeon® processor E7-8890 v3 platform supports 1.66 times
more concurrent users of our software compared to the previous processor
generation. This enables more physicians to simultaneously access the
critical imaging information they need, when and where they need it.”
Tim Dawson – Senior Director of Software Engineering, Vital Images, Inc.
 Vital’s solutions provide rich clinical tools for viewing human anatomy
in 2D, 3D and 4D for efficient and effective patient care.
 Higher user density enabled by increased core count, memory capacity
and memory performance.
 Lower administrative costs than separate workstations.
 A single system can support more simultaneous users, which may
increase productivity and improve patient care.
www.vitalimages.com
1 - Testing conducted on Vital Images VitreaAdvanced* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Vital Images. For complete testing configuration details, see SLIDE 88.
Up to
66%
faster!!
Workload: VitreaAdvanced* software clinical applications involving
3D image processing algorithms and 3D volumetric rendering.
Vital Images VitreaAdvanced* increased
concurrent users supported from 12 to 20, 1.66X,
Health & Life Sciences
New possibilities for larger radiological practices with a reduced
infrastructure footprint and cost of operations!

57
Scripps Translational Science*
DNA Sequencing Pipeline*
“Intel® Xeon® processor E7-8890 v3 helps run DNA sequencing pipelines up
to 60% faster over previous generation for Whole exome datasets that are
essential to accelerate the adoption of genomic medicine in the industry.
Additionally the Intel® Solid State Drive helps reduce the I/O bottleneck
while dealing with large datasets. ”
Ali Torkamani – Director of Genomics Informatics and Drug Discovery,
The Scripps Translational Science Institute
Up to
60%
faster
Intel technologies are a catalyst to accelerating Genomic
Medicine!
1 - Testing conducted on Scripps* DNA sequencing pipelines comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Scripps. For complete testing configuration details,
see SLIDE 88. Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific
computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully
evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit http://www.intel.com/performance.
 Faster DNA sequencing is an industry necessity to drive Genomic
Medicine/Personalized Medicine.
 Quick processing of large amounts of genomics data helps accelerate
scientific research.
 In trio studies, samples from parents, siblings or other family members
need to be processed, meaning large amounts of data processing
requires Intel® Xeon® processor technologies to provide insight into
data for patient conditions.
www.stsiweb.org
0
1
Up to
60%
faster!
Workload: Pipeline of software algorithms that help align,
recalibrate the DNA sequences to produce an output that could be
used for annotation and interpretation of genes.
Scripps* DNA sequencing processing time
improved from 111 to 69 minutes, 1.60X, with 4S

0
1
58
GROMACS*
GROMACS is a versatile package to perform molecular dynamics. It is
primarily designed for biochemical molecules like proteins, lipids and
nucleic acids that have a lot of complicated bonded interactions. And
because GROMACS is extremely fast at calculating the non-bonded
interactions (that usually dominate simulations), many groups are also
using it for research on non-biological systems, e.g. polymers.
More simulations per day and more atoms in a box!
1 - Testing conducted on GROMACS* comparing Intel® Xeon® Processor E7-8890 v3 with Intel® Xeon® Processor E7-4890 v2. Testing done by Intel. For complete testing configuration details, see SLIDE 88.
 GROMACS increases performance because of more cores, more threads,
and more memory with the Intel® Xeon® processor E7-8890 v3.
 Intel Xeon processor E7-8890 v3 delivers faster speed and higher
throughput, enabling support of more users.
www.gromacs.org
Up to
60%
faster!
Workload: Two representative public workloads. The first one is
141K atoms computed with PME method, the second one is 3.1M
atoms computed with RF method.
GROMACS* increased concurrent users support
from 1.656 to 2.661, 1.60X, with 4S Intel® Xeon®

0
1
59
Epic*
“Moving from the Intel® Xeon® processor E7-4890 v2 to Intel® Xeon®
processor E7-8890 v3, we saw the scalability of a single operational
database server increase over 40%. With these gains, we expect our
customers to scale further with a smaller datacenter footprint and lower
TCO.”1
Seth Hain - Senior Performance Engineer at Epic
 Performance increase of up to 42% for scale up Epic* SMP
deployments with Intel Xeon processor E7-8890 v3 compared to the
previous processor generation.
 Increased Intel Xeon processor E7-8890 v3 scalability means that even
larger Epic customers can now implement their Epic EMR systems with
confidence on Intel Xeon processors.
 Increased scalability brings reduced infrastructure costs and higher ROI
to achieve organizational objectives while providing excellent patient
and clinician response times.
 The Intel Xeon processor E7-8890 v3 increased memory bandwidth,
cores and IO, allowing for highly scalable platforms.
www.epic.com
1 - Testing Cache 2015.1, comparing 4S Intel® Xeon® Processor E7-8890 v3 to 4S Intel® Xeon® Processor E7-4890 v2. Testing performed by Epic measuring database performance. For complete configuration details, see SLIDE 88.
Up to 42% greater performance and significantly reduced TCO !
Epic* increased database accesses per second from
~4.9M to ~7M, up to 1.42X, with the 4S Intel® Xeon®
Up to
42%
faster!!

0
1
60
Dassault Systèmes
BIOVIA Pipeline Pilot
“Intel’s new Intel® Xeon® processor E7-8890 v3 streamlines large scale Next
Generation Sequencing (NGS) data analysis, reducing the challenges of IT
departments in the scientific community that are burdened with traditional
cumbersome HPC clusters.”
Tim Moran - Director Life Science Research Marketing, Dassault Systèmes
www.3ds.com
1 - Testing conducted on Dassault Systemes BIOVIA* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Intel. For complete testing configuration details, see SLIDE 89.
Up to
18%
faster!
Workload: Bowtie 2 human genome mapping.
BIOVIA Pipeline Pilot* improved BWA mapping
time (hours) from 4:57 to 4:11, 1.18X, with 4S
 Built on the industry-leading BIOVIA Foundation*, the Pipeline Pilot
Next Generation Sequencing* (NGS) Collection offers a comprehensive
set of NGS data analysis pipelines for easy analysis and interpretation
of the massive datasets.
 Next gen sequencing on latest and faster architectures helps scientific
research in drug development by enabling the current analysis run
faster.
Pipeline Pilot* Improved efficiency and enables IT departments
to spend more time on Innovation

TechnicalComputing
62
Health
& Life Sciences
Up to
Faster1
25%
Financial
Services
Big Data
Analytics
Core Business
& Telecom
Enterprise
Database
NO COMPROMISE ARCHITECTURE
The Intel® Xeon® processor E7 v3 family
delivers major gains in performance, density,
and energy-efficiency, to support larger
models and faster iterations, all at lower cost.
Whether it’s advanced simulation and
modeling applications that transform the way
companies create and test new product
designs or faster imaging for more efficient
drug discovery and deeper genomic analysis by
health care organizations and bio-medical
research teams, a new generation of Intel
products is helping to increase the speed and
performance of vital applications.
1 – Demonstrated with Mentor Graphics* proof point.

63
Mentor Graphics*
Calibre nmDRC*
“The Intel® Xeon® processor E7-8890 v3 offers an average of 1.15X and a
peak of 1.25X multi-job throughput increase over the previous processor
generation running Calibre nmDRC*. Calibre nmDRC is highly-scalable and
is the IC industry leader for design rule checking and physical verification.”
Michael White – Director of Product Marketing, Calibre Physical
Verification Products, Mentor Graphics
www.mentor.com
Up to
28%
faster
Mentor Graphics industry-leading Calibre nmDRC* product sees
higher throughput gain up to 1.25X!
1 - Testing conducted on Mentor Graphics Calibre* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Mentor Graphics. For complete testing configuration details, see SLIDE 89.
Technical Computing
0
1
Mentor Graphics Calibre nmDRC* improved
throughput with 4S Intel® Xeon® processor
E7-8890 v3
Up to
25%
faster!
Workload: Design Rule Check.
 The physical verification tool used by every major foundry to develop
these design rules is Calibre nmDRC.

0
1
ANSYS*
CFX* 16.0
“Great memory and socket scalability combined with more operations per
clock is allowing our customers to achieve cluster performance running CFX*
on just a single Intel® Xeon® Processor E7-8890 v3 server node.“1
Dan Williams - Senior CFX Development Manager, ANSYS, Inc.
64
www.ansys.com
 ANSYS CFX* is a general purpose fluid dynamics program. The Intel Xeon
Processor E7-8890 v3 delivers great scalability and more operations per
clock, allowing standard servers to achieve levels of performance
previously associated with supercomputers.
 For end-users, 1.21x improved performance compared to the previous
generation processor means faster turnaround or solving larger tasks with
the same resources, and with lower TCO.
Technical Computing
Do more with the same resources and lower your TCO!
Up to
21%
faster!
Workload: CFX 16 AirFoil, LeMansCar, Pump.
ANSYS CFX* improved geomean processing time
from 32.92 to 27.07 seconds, 1.21X, with 4S
1 - Testing conducted on ANSYS CFX 16* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Intel. For complete testing configuration details, see SLIDE 89.

0
1
65
Altair*
RADIOSS*
“With the Intel® Xeon® processor E7-8890 v3, we were able to achieve
outstanding performance running Altair HyperWorks RADIOSS* solver. With
72 cores in a 4-socket machine, plus improvements in memory capacity,
speed and control, this new processor sets a new standard for HPC on both
small and large systems. Our RADIOSS customers will be very happy with
the jump in performance made possible by this new release.”
Eric Lequiniou - HPC Director, Altair
 RADIOSS*, a widely used application for crash simulation, takes full
advantage of more cores, more threads, and more memory of the Intel®
Xeon® processor E7-8890 v3.
 The Intel Xeon processor E7-8890 v3 provides a continually increasing
level of scalability for Altair customers, allowing them to run very complex
simulations with amazing performance on a single node, instead of a
cluster.
www.altair.com
Altair RADIOSS* improved performance time from
276 to 229 seconds, 1.2X, with 4S Intel® Xeon®
1 - Testing conducted on Altair Radioss* comparing 4S Intel® Xeon® Processor E7-8890 v3 with 4S Intel® Xeon® Processor E7-4890 v2. Testing done by Intel. For complete configuration details, see SLIDE 89.
Up to
20%
faster!
Workload: NEON 1M 8ms DP.
Run complex simulations faster with more cores & greater
memory capacity!
Technical Computing

SolutionsthatWorkBetterTogether
67
HardwareUpgrade
Intel® Xeon® Processor E7-8890 v3
Scalability
• Increased parallelism - up to 18 cores, up to 8-
sockets per server natively
• Up to 1.5 TB supported per socket of DDR3 or DDR4
delivering up to 3.6X more memory bandwidth1
Reduce performance bottlenecks with Intel® SSD Data
Center Family
Baseline
• Microsoft SQL Server* 2008 R2
• Intel® Xeon® Processor E7-4870
2011 SoftwareUpgrade
Microsoft SQL Server* 2014
Performance and Scalability enhancements:
• In-Memory Online Transaction Processing
• Faster query response times
2015
2015
BetterTogether
Upto2.2x
performance improvement2
Microsoft* and Intel® deliver
outstanding reliability and speed
for online transaction processing
Workload: Brokerage firm OLTP
LatestMicrosoft
SQLServer*
LatestIntel® Xeon® processor
E7Family
+
SoftwareandworkloadsusedinperformancetestsmayhavebeenoptimizedforperformanceonlyonIntel
microprocessors.Performancetests,suchasSYSmarkandMobileMark,aremeasuredusingspecificcomputer
systems,components,software,operationsandfunctions.Anychangetoanyofthosefactorsmaycausetheresults
tovary.Youshouldconsultotherinformationandperformanceteststoassistyouinfullyevaluatingyour
contemplatedpurchases,includingtheperformanceofthatproductwhencombinedwithotherproducts.Formore
informationgotohttp://www.intel.com/performanceConfigurations(seenextslideformoredetails)
Footnotescanbefoundonthefollowingslide..*
Copyright©2015IntelCorporation.Allrightsreserved.
=

ConfigurationDetails
Microsoft SQL Server* and Intel® Xeon® Processor E7 v3 Family
68
Processor Intel® Xeon® Processor E7-4870 Intel® Xeon® Processor E7-8890 v3
Sockets 4 4
Memory 512 GB (64X8 GB DDR3LV-1067MHz) 512GB (16 x 8GB DDR4-2133)
Disk Intel® SSD X25-E Series Intel® SSD DC S3700 Series
OS Distribution Microsoft Windows Server 2008 R2 Microsoft Windows Server 2012
Software Microsoft SQL Server 2008 R2 Microsoft SQL Server* 2014
OLTP Throughput
(ktps)
2.8 6.259
Source TR 1121 TR 41
1. Up to 3.6x memory bandwidth based on STREAM(triad) benchmark comparing baseline Intel® Server platform with four Intel® Xeon® Processor X7560, 64x4GB DDR3-1066 scoring 71.578 GB/sec (source: Intel® technical report #1129) to the new Intel®
Server platform with four Intel® Xeon® Processor E7-8890 v3, 64x16GB DDR4-1600 scoring 269.7 GB/sec. (source: Intel® technical report #106).
2 Up to 2.2x performance based on SQL Server* OLTP workload comparing baseline Quanta QSSC-S4R* platform with four Intel® Xeon® Processor E7-4870 using SQl Server 2008 R2 scoring 2.8 ktps (source: Intel technical report #1121) to the new
Lenovo System x3850 X6* platform using SQL Server 2014 scoring 6.964 ktps (source: Intel® technical report #41) Copyright © 2015 Intel Corporation. All rights reserved.

69
HardwareUpgrade
Scalability
Center Family
Baseline
• Microsoft SQL Server* 2005
Enterprise Edition x64 SP2
• Microsoft Windows Server*
2003
• Intel® Xeon® Processor X7350
2007
SoftwareUpgrade
2015
2015
BetterTogether
Upto16x
Moretransactionsper
secondat1/6th thecostper
transaction2
Workload: TPC-E*
LatestMicrosoft
SQLServer*
E7Family
+
informationgotohttp://www.intel.com/performanceConfigurations(seenextslideformoredetails):
Footnotescanbefoundonthefollowingslide..*
Microsoft* and Intel® deliver
amazing costs savings while
increasing transaction speed
=
Microsoft SQL Server 2005
& Microsoft Windows
Server 2003 EOL

70
Processor Intel® Xeon® Processor X7350 Intel® Xeon® Processor E7-8890 v3
Sockets 4 4
Memory 128 GB 512GB (16 x 8GB DDR4-2133)
Disk 15K 4Gbps FC E-DDM Hot-Swap HDD Intel® SSD DC S3700 Series
OS Distribution Microsoft Windows Server 2003 Microsoft Windows Server 2012
Software
Microsoft SQL Server 2005 Enterprise Edition
x64 SP2
tpsE
$/tpsE
419
$1527.25/tpsE
6964.75
$245.98/tspE
Source Source available 12/07/07 Source available 07/31/2015
1. Up to 3.6x memory bandwidth based on STREAM(triad) benchmark comparing baseline Intel® Server platform with four Intel® Xeon® Processor X7560, 64x4GB DDR3-1066 scoring 71.578 GB/sec (source: Intel® technical report #1129) to the new
Intel® Server platform with four Intel® Xeon® Processor E7-8890 v3, 64x16GB DDR4-1600 scoring 269.7 GB/sec. (source: Intel® technical report #106).
2. Up to 16x performance improvement at as low as 1/6th cost claim based on TPC-E* workload, comparing baseline IBM System x3850* M2 running Microsoft SQL Server* 2005 with four Intel® Xeon® processor X7350 2.93GHz (4 processors, 64 cores,
64 threads) Historical Result: costing $639,917, 419 tpsE, $1527.25/tpsE (http://www.tpc.org/4004) to the new Lenovo x3850 X6* system running Microsoft SQL Server* 2014 with four Intel® Xeon® processor E7-8890 v3 2.5GHz (4 processors, 72
cores, 144 threads) costing $1,723,004, 6964.75 tpsE, $245.98/tpsE (http://www.tpc.org/4072).

71
HardwareUpgrade
Scalability
Center Family
Baseline
• Microsoft SQL Server* 2008 R2
2010 SoftwareUpgrade
2015
2015
BetterTogether
Upto6.8x
performanceat1/20th cost
perQuery2
Delivers deeper analytics
and business insight through
a modern, fully-integrated
enterprise-class data platform
Workload: TPC-H* @ 3000 GB
LatestMicrosoft
SQLServer*
E7Family
+
Footnotescanbefoundonthefollowingslide..
=

72
Sockets 16 4
Memory 512 GB (64X8 GB DDR3LV-1067MHz) 512GB (16 x 8GB DDR4-2133)
Disk Intel® SSD X25-E Series Intel® SSD DC S3700 Series
OS Distribution Microsoft Windows Server 2008 R2 Microsoft Windows Server 2012
Software Microsoft SQL Server 2008 R2 Microsoft SQL Server* 2014
TPC-H* QphH @ 3000GB
$/QphH @ 3000GB
102,778
$21.04/QphH @ 3000GB
700,392
$0.98/QphH @ 3000GB
Source Source available 5/06/10 Source available 05/26/15
2. Up to 6.8x performance improvement and up to 20x lower cost claim based on TPC-H workload comparing baseline Unisys ES7000* Model 7600R Enterprise Server (16s) with four Intel® Xeon® processor x7560 2.26GHz (4 processors, 8 cores, 16
threads) Historical Result: scoring 102,778 QphH, costing $2,162,895, $21.04/QphH (source: http://tpc.org/3245) to the new Intel® Server platform with four Intel® Xeon® E7-8890 v3 2.5GHz (4 processors, 72 cores, 144 threads) scoring 700,392
QphH costing $691,524, $0.98/QphH (Source: http://www.tpc.org/3313)

LatestMicrosoft
WindowsServer*
LatestIntel®Xeon®processor
E7Family
+ SolutionsthatWorkBetterTogether
73
HardwareUpgrade
Scalability
• Up to 1.5 TB supported per socket of DDR3 or
DDR4 delivering up to 3.6X more memory
bandwidth1
Up to 1.9X performance improvement3
Baseline
• Windows Server 2008 R2
• Intel® Xeon® Processor E7-4870
2011
SoftwareUpgrade
Microsoft Windows Server* 2012 R2
Support up to 640 logical processors
to massively increase parallelizability
Support up to 4TB memory
Up to 7% performance improvement4
2015
2015
BetterTogether
Upto2x
Performanceimprovement2
Microsoft and Intel together
provide a powerful platform for
server Java applications
Benchmark: Server Side Java Workload
Copyright©2015IntelCorporation.Allrightsreserved..
=

Microsoft Windows Server* and Intel® Xeon® Processor E7 v3 Family
74
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors.
Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other
information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit http://www.intel.com/performance.
2. Up to 2x performance based on Server Side Java* workload comparing baseline Intel® Emerald Ridge platform with four Intel® Xeon® Processor E7-4870 using Windows Server 2008 R2 scoring 58,054 to the new Intel Server platform with four Intel®
Xeon® Processor E7-8890 V3 using Windows Server* 2012 R2 scoring 116,560.
3. Up to 1.9x performance based on Server Side Java* workload comparing baseline Intel® Emerald Ridge platform with four Intel® Xeon® Processor E7-4870 using Windows Server 2008 R2 scoring 58,054 to the new Intel Server platform with four Intel®
Xeon® Processor E7-8890 V3 using Windows Server* 2008 R2 scoring 108,789.
4. Up to 7% performance improvement based on Server Side Java* workload comparing baseline Intel® Server platform with four Intel® Xeon® Processor E7-8890 V3 using Windows Server* 2012 R2 scoring 108,789 to Intel® Server platform with four
Intel® Xeon® Processor E7-8890 V3 using Windows Server* 2012 R2 scoring 116,560.
Copyright © 2015 Intel Corporation. All rights reserved.
Processor
Intel® Xeon® Processor
(formerly Westmere) E7-4870
(formerly Westmere) E7-4870
(formerly Haswell) E7-8890 v3
(formerly Haswell) E7-8890 v3
(18C, 2.3 GHz, 145W)
Sockets 4 4 4 4
Memory
256 GB
(64x4 GB DDR3-1067)
256 GB
(64x4 GB DDR3-1067)
512GB
(16 x 8GB DDR4-2133)
512GB
(16 x 8GB DDR4-2133)
OS Distribution Windows Server 2008 R2 Windows Server 2012 R2 Windows Server 2008 R2
Windows Server 2012 R2
Ops Performance
(Server Side Java)
55,724 58,054 108,789 116,560

75
HardwareUpgrade
Up to 3X2 more performance from hardware upgrade
• Up to 1.5 TB supported per socket of DDR3 or
DDR4 delivering up to 3.6X more memory
bandwidth1
Baseline
• Red Hat Enterprise Linux* 6.1
• Intel Compiler 12.0
2010
SoftwareUpgrade
Red Hat Enterprise Linux* 7.1 w/ Intel Compiler 15
• Performance Management suite to tune and optimize
performance to selected system profile
• Automatic NUMA balancing
2015
2015
BetterTogether
Upto3.3x
HigherPerformance4
Red Hat* and Intel® continue to
propel the industry standard for
general purpose computing to new
heights
Benchmark: Estimated SPECint*_rate_base2006
LatestRedHat
EnterpriseLinux*
E7Family
+
Footnotescanbefoundonthefollowingslide.
=

Red Hat Enterprise Linux* and Intel® Xeon® Processor E7 v3 Family
76
Processor Intel® Xeon® Processor X7560
Intel® Xeon® Processor E7-8890
v3
Intel® Xeon® Processor E7-8890
v3
Sockets 4 4 4
Memory
1TB
(64x16 GB DDR3-1066MHz)
1 TB
(64x16 GB DDR4-1600MHz)
1 TB
(64x16 GB DDR4-1600MHz)
OS Distribution Red Hat* Enterprise LINUX 6.1 Red Hat* Enterprise LINUX 6.5 Red Hat* Enterprise LINUX 7.1
Compiler Intel® Compiler 12.0.1.116 Intel® Compiler 12.0 Intel® Compiler 15.0
Performance
(SPECint_rate_
base2006)
802 (Intel Est.) 2460 (Intel® Est.) 2,670 (Intel Est.)
2. Up to 3x performance based on estimated SPECint*_rate_base2006 workload comparing baseline Cisco UCS C460 M1 platform with four Intel® Xeon® Processor X7560 using Red Hat Enterprise Linux 6.1 scoring 802 (source) to the new Intel Server
platform with four Intel® Xeon® Processor E7-8890 V3 using Red Hat Enterprise Linux 6.1 scoring 2460 (source: Intel technical Report #245)
3. Up to 3.3x performance based on estimated SPECint*_rate_base2006 workload comparing baseline Cisco UCS C460 M1 platform with four Intel® Xeon® Processor X7560 using Red Hat Enterprise Linux 6.1 scoring 802 (source) to the new Intel
Server platform with four Intel® Xeon® Processor E7-8890 V3 using Red Hat Enterprise Linux 7.1 scoring 2670 (source: Intel technical Report #220)

77
HardwareUpgrade
• Up to 18 cores, up to 8-sockets per server natively
• Up to 1.5 TB of DDR3 or DDR4 delivering up to 3.6X
more memory bandwidth2
• NEW Intel® TSX which boosts performance of
multi-threaded application when sharing critical
data in the memory
Baseline
• SAP HANA* 1 SPS 8
• Intel® Xeon® Processor
E7-4890 v2
2012
SoftwareUpgrade
SAP HANA* 1 SPS 9
• 1.8X performance improvement1
• Dynamic data tiering capability
• Hadoop Map Reduce integration
2015
2015
BetterTogether
Upto6x
Performanceimprovement3
Intel and SAP simplify IT by
delivering a purpose-built
processor and platform
combination to reduce complexity
and optimize performance
Benchmark: insert and select OLTP with Intel TSX
latestSAPHANA*
E7Family
+
=

SAP HANA* and Intel Xeon Processor E7 v3 Family
78
Processor
(15 Cores, 2.8 GHz, “Ivy Bridge-EX”)
(15 Cores, 2.8 GHz, “Ivy Bridge-EX”)
(18 Cores, 2.5 GHz, “Haswell-EX”)
Sockets 4 4 4
Memory 512 GB 512 GB 512 GB
OS SuSE* Linux Enterprise Server 11SP3 SuSE* Linux Enterprise Server 11SP3 SuSE* Linux Enterprise Server11SP3
Software SAP HANA* 1 SPS 8 SAP HANA* 1 SPS 9 SAP HANA* 1 SPS 9
OLTP Transactions
per minute
14,327 26,139 89,619
Source Intel® Internal Testing Intel® Internal Testing Intel® Internal Testing
1. Up to 1.8x performance based on internal OLTP workload comparing Intel Server platform with four Intel® Xeon® Processor E7-4890 v2 using SAP HANA* 1 SPS 8 scoring 14,327 to Intel Server platform with four Intel® Xeon® Processor E7-4890 v2
using SAP HANA* 1 SPS 9 scoring 26,139.
2. 2. Up to 3.6x memory bandwidth based on STREAM(triad) benchmark comparing baseline Intel® Server platform with four Intel® Xeon® Processor X7560, 64x4GB DDR3-1066 scoring 71.578 GB/sec (source: Intel® technical report #1129) to the new
3. Up to 6x performance based on SAP OLTP internal insert and select tests measuring transactions per minute comparing Intel Server platform with four Intel® Xeon® Processor E7-4890 v2 using SAP HANA* 1 SPS 8 scoring 14,327 to the new Intel
Server platform with four Intel® Xeon® Processor E7-8890 V3 using SAP HANA* 1 SPS 9 scoring 89,619.

79
HardwareUpgrade
Scalability
• Reduce performance bottlenecks with Intel® SSD
Data Center Family
Baseline
• VMware vSphere* (ESXi) 4.1
2012
SoftwareUpgrade
VMware vSphere* (ESXi) 6
• Increase in host size
• 3X increase in CPUs per host (160 to 480)
• 12X increase in RAM per cluster (1TB to 12TB)
2015
2015
BetterTogether
Upto17xImprovementin
VirtualizationPerformance
atupto1/5th costperVM2,3
VMware takes advantage of new
Intel networking and storage
opportunities to deliver
outstanding virtual machine
performance
Benchmark: Server Virtualization Consolidation workload
LatestVmWAREVSPHERE*(ESXi)
E7Family
+
Softwareandworkloadsusedinperformancetestsmayhavebeenoptimizedforperformanceonlyon
Intelmicroprocessors.Performancetests,suchasSYSmarkandMobileMark,aremeasuredusingspecific
computersystems,components,software,operationsandfunctions.Anychangetoanyofthosefactors
maycausetheresultstovary.Youshouldconsultotherinformationandperformanceteststoassistyou
infullyevaluatingyourcontemplatedpurchases,includingtheperformanceofthatproductwhen
combinedwithotherproducts.Formoreinformationgotohttp://www.intel.com/performance
Configurations(seenextslideformoredetails):
Footnotescanbefoundonthefollowingslide.
=

VMware ESXi* and Intel® Xeon® Processor E7 v3 Family
80
Sockets 4 4
Memory 192GB (24 x 8GB DDR2-667) 1TB (64 x 16GB DDR4-1600)
Storage Seagate* SAS 15K HDD Series Intel® SSD Datacenter Family
VMWare Software
Distribution
VMWare vSphere* (ESXi) 4.1 VMWare vSphere* (ESXi) 6
Relative Virtualized
Performance
94.28 @ 6 VMs 1603 @ 90 VMs
Source Intel Internal Technical Report #254 Intel Internal Technical Report #232
2. Up to 17x throughput gain vs. 5-year old servers for VM consolidation claim using internal virtualized consolidation workload scoring application throughput score @ given number of VMs. Configurations include: 1) 1-Node, 4 x Intel® Xeon® Processor
X7460 with 192 GB Total Memory on VMware ESXi* 4.1 Update 2 using Guest OS RHEL6.4-64bit. Data Source: Request Number: 1392. Scoring 94.28 @ 6 VMs. 6) 1-Node, 4 x Intel® Xeon® Processor E7-8890 v3 with 1024 GB 1024 GB on VMware ESXi* 6
Update 1 using Guest OS RHEL6.4-64bit. Data Source: Request Number: 232. Scoring 1603 @ 90 VMs.
3. Up to 1/5th the cost per VM based on estimated hardware costs including acquisition costs, memory, storage and networking (no S/W) of the old system at $31K ($5166/VM) to the new configuration at $101.5K ($1127/VM). (source: Intel technical
Report 232)

platformsupportWhere to go for more information about the Intel® Xeon® processor E7-8800/4800 v3
81

helpfullinks
 Intel® Xeon® processor E7 v3 family:
• Product page: http://www.intel.com/XeonE7
• Support page: http://www.intel.com/p/en_US/support/highlights/processors/xp-e7-8800
• Software Solutions page: www.intel.com/XeonE7SoftwareSolutions
• Data Center Performance page: http://www.intel.com/performance/datacenter
• World Records page: www.intel.com/E7v3records
 Intel® Solid-State Drive Data Center P3700:
• Product Page: http://www.intel.com/content/www/us/en/solid-state-drives/intel-ssd-dc-family-for-pcie.html
• Support Page: http://www.intel.com/p/en_US/support/highlights/ssdc/ssd-dc-p3700
 Intel® Ethernet Solutions:
• Product Page: http://www.intel.com/content/www/us/en/network-adapters/converged-network-adapters.html
• Support Page: http://www.intel.com/p/en_US/support/highlights/network/xl710
 Intel® Server Software for Code Vectorization, Parallelism and Optimization:
• Product Page: https://software.intel.com/en-us/intel-mkl
• Product Page: https://software.intel.com/en-us/intel-parallel-studio-xe
82

84
ConfigurationDetails:Intel®Xeon®ProcessorE7v3FamilyKeyGenerationalPerformanceIncreases(slide10)
1 - Up to 39% more OLTP database transactions claim based on OLTP warehouse workload measuring transactions per minute (tpm) using Oracle* 11g R2 on Red Hat Enterprise Linux 6.5 – internal technical
reports (TR) #235, #40. Configurations:
BASELINE: 4S Intel® Xeon® processor E7-4870 using DDR3-1333 memory, 60x 800GB Intel DC S3700 SSDs. Score: 4789K tpm.
NEW: 4S Intel Xeon processor E7-8890 v3 using 2 TB DDR4-1600 memory, 84x 800GB Intel DC S3700 SSDs scoring 6602K tpm.
2 - Up to 6x more in-memory OLTP transactions with Intel® TSX using SUSE* Linux* Enterprise Server 11 SP3 running SAP internal OLTP Insert and select tests measuring transactions per minute (TPM).
Configurations:
BASELINE: 4S Intel® Xeon® processor E7-4890 v2, 512 GB memory, SUSE Linux Enterprise Server 11 SP3, SAP HANA* 1 SP8 scoring 14,327 tpm.
NEW: 4S Intel Xeon processor E7-4890 v2, 512 GB memory, SUSE Linux Enterprise Server 11 SP3, SAP HANA 1 SP9 with Intel TSX enabled scoring 89,619 tpm.
3 - Up to 70% claim based on SAS Mixed Analytics workload measuring sessions per hour using SAS Business Analytics* 9.4 M2 on Red Hat* Enterprise LINUX 7. Configurations:
BASELINE: 4S Intel Xeon processor E7-4890 v2, 512 GB DDR3-1066 memory, 4x 2.0 TB Intel DC P3700 SSDs + 8x 800 GB Intel DC S3700 SSDs scoring 0.11 sessions/hour.
NEW: 4S Intel Xeon processor E7-8890 v3, 512 GB DDR4-1600 memory, 4x 2.0 TB Intel DC P3700 SSDs + 8x 800 GB Intel DC S3700 SSDs scoring 0.19 sessions/hour.
4 - Up to 39% application throughput for VM consolidation claim based on internal virtualized consolidation workload, running VMware ESXi*5.5 measuring throughput (min. QOS), with 2x Intel® RAID
Controller RS2P1008, 2x Intel 82599EB 10GbE – internal technical reports (TR) #131, #186. Configurations:
BASELINE: 4S Intel Xeon processor E7-4890 v2, RHEL* 6.2, 1 TB DDR3-1333 memory, 2x LSI and 2x NewSYS enclosures, 18x 400GB Intel SATA SSDs, 16x 300GB SSDs, 1x Intel® 82576 Dual-port GbE; Scoring
1150 @ 63 VMs.
NEW: 4S Intel Xeon processor E7-8890 v3, RHEL 6.4, 1 TB DDR4-1600 memory, 4x LIS 620J, 48x 64GB X25-E Intel SATA SSDs, 24x 100GB Intel 710 SATA SSDs, 24x 800GB Intel DC S3700 SSDs, 1 HP
ProCurve 6400cl; Scoring 1606 @ 90 VMs.
Intel®Xeon®ProcessorE7v3FamilySoftwareplatformvalueproposition(slide12)
1 - Claim based on SAS* Mixed Analytics workload measuring sessions per hour using SAS Business Analytics* 9.4 M2 on Red Hat* Enterprise Linux* 7. Configurations:
BASELINE: 4S Intel® Xeon® processor E7-4890 v2, 512 GB DDR3-1066 memory, 16x 800 GB Intel® SSD DC S3700, scoring 0.11 sessions/hour.
NEW: 4S Intel® Xeon® processor E7-8890 v3, 512 GB DDR4-1600 memory, 4x 2.0 TB Intel® SSD DC P3700 SSDs + 8x 800 GB Intel® DC S3700 SSDs, scoring 0.19 sessions/hour. See
http://www.intel.com/performance/datacenter for more details.
2 - Claim based on SAP* OLTP internal insert and select tests measuring transactions per minute (tpm) on SUSE* LINUX Enterprise Server 11 SP3. Configurations:
BASELINE: 4S Intel Xeon processor E7-4890 v2, 512 GB memory, SUSE* LINUX Enterprise Server 11 SP3, SAP HANA* 1 SP8 scoring 14,327 tpm.
NEW: A) 4S Intel Xeon processor E7-4890 v2, 512 GB memory, SUSE* LINUX Enterprise Server 11 SP3, SAP HANA* 1 SP9 scoring 26,139 tpm.
B) 4S Intel Xeon processor E7-8890 v3, 512 GB memory, SUSE* LINUX Enterprise Server 11 SP3, SAP HANA* 1 SP9 – Intel TSX disabled scoring 39,330 tpm. C) 4S Intel Xeon processor E7-8890 v3, 512 GB
memory, SUSE* LINUX Enterprise Server 11 SP3, SAP HANA* 1 SP9 – Intel TSX enabled scoring 89,619 tpm. See http://www.intel.com/performance/datacenter for more details.
3 - Claim based on estimated SPECint_rate_base2006* results assuming a scenario of 100 each 5-year old 4-socket servers using Intel Xeon processor X7460 scoring 274 each compared to 11 each
comparable Intel Xeon processor E7-8890 v3-based servers scoring 2650 each. The 23-month ROI payback is based on an initial investment of $645,786 with a net savings of $742,126 for a 115% ROI and
40% IRR when purchasing each server at an estimated street price of $56,228 for with $1505 installation and disposal cost per server providing 81% savings of power/cooling and lower management costs with
89 fewer deployed servers in this scenario. See http://www.intel.com/performance/datacenter and http://estimator.intel.com/serverroi/ for more details.

85
ProofpointConfigurationDetails:
Ayasdi Core* 15: This analytic benchmark emulates the behavior of our appliance under load. This is a modified version of the standard benchmark, which our QA team uses to validate new releases. The
benchmark was scaled down to accommodate the small size of the test cluster size and limited memory resources. Performance is measured in terms of 1) the end-to-end latency of the analyses, and 2) the
overall throughout of analyses that can be supported by the cluster, with acceptable latency.
BASELINE: RHEL 7, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 1 TB DDR3-1600, SAS disk for OS, Intel® SSD P3700 for data, 1GB network speed
NEW: RHEL 7, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 1 TB DDR4-1600, Intel® AVX2, SAS disk for OS, Intel® SSD P3700 for data, 1GB network speed. Testing done by Ayasdi 3/25/2015
SAS* 9.4 M2: The new SAS* mixed analytic workload represents the type of SAS processing that most SAS customers use. This workload created a multi-user benchmarking scenarios to simulate the workload
of a typical Foundation SAS or Base SAS / STAT customer. The goal of these scenarios is to evaluate the multi-user performance of SAS on various platforms. Various sized mixed analytic workloads were
created to simulate many users utilizing CPU, RAM and I/O resources which SAS programs heavily utilize during typical program execution. The test was designed to mimic this workload and be scaled to
represent different sized customer environments. In this specific case, 60 sessions of mixed workload are used to measure the performance for platform comparison between Intel® Xeon® processor E7-8890 v3
and Intel Xeon processor E7-4890 v2. It runs a 60 simultaneous job workload of a repetitive mix of jobs.
BASELINE: RHEL 7, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512 GB DDR3-1066, 16 Intel® SSD S3700, 1GB network speed
NEW: RHEL 7, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 512 GB DDR4-1600, 4 Intel® SSD P3700, 8 Intel® SSD S3700, 1GB network speed. Testing done by Intel 2/20/2015
Software AG Apama*: The workload is a micro-operation that is performed very often inside Apama. A running Apama correlator (the event engine) is composed of a number of event processing pipelines, that
work together. The workload receives events in the pipeline stage and publishes an output from the Apama correlator.
BASELINE: RHEL 6.5, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512 GB DDR3-1333, SATA disk, 10GB network speed
NEW: RHEL 6.5, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 512 GB DDR4-1600, SATA disk, 10GB network speed. Testing done by Software AG 3/30/2015
Nomura Research Institute BOS*: Real-time Stock calculation (Normalizedd TPC-H benchmark schema with Bill of Materials). Deploy a Java Virtual Machine and two PostgreSQL database instances for each
NUMA node. Data Manipulation on top of extremely Normalizedd database schema (INSERT ONLY design constraint without UPDATE and DELETE operations).
BASELINE: RHEL 7, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 256 GB DDR3-1333, 2 SAS disks, 10GB network speed
NEW: RHEL 7, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 512 GB DDR4-1600, 2 SAS disks, 10GB network speed. Testing done by Nomura Research Institute 2/10/2015
IBM InfoSphere DataStage:* Typical Data Warehousing Integration Scenario; based on Brokerage Firm Business Scenario that performs extract, transform and load operations (ETL). CPU intensive ETL test
(DimTrade) used. Data volume size is 53GB.
BASELINE: RHEL 7, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128 GB DDR3-1333, 2 SATA disks, 1GB network speed
NEW: RHEL 7, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 258 GB DDR4-1600, 2 SATA disks, 1GB network speed. Testing done by IBM 3/3/2015
IBM Informix* 12.10: Analytical queries on a large set of sensor time series data, which originates from publicly available U.S. government data (NWIS - National Water Information System). The data and query
workload is typical for better understanding environment and climate changes over prolonged periods of time. The NoSQL data used in the workload represents periodic measurements (in 15 minute intervals)
from more than 800 metering points over a period of roughly 100 years. This amounts to a total of more than 2.5 billion data records, each comprising the timestamp and multiple measurement values. The
analytics workload performed on this data represents various aggregations typically done on time series data. Workload comprises of 2 queries.
BASELINE: RHEL 6.5, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 1 TB DDR3-1333, 2 SATA disks for OS, 2 Intel® SSD 910 for data, 1GB network speed.
NEW: RHEL 6.5, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 1 TB DDR4-1600, 2 SATA disks for OS, 2 Intel® SSD 910 for data, 1GB network speed. Testing done by IBM 3/5/2015
Neusoft Aclome* 4.7: The test uses their load test tool to emulate the workload pressure brought by end user requests to calculate the throughput and response time of the system in a scale of production
environment. The main request type is report generation, which is considered a heavy workload request. The performance is measured by the finished requests per second from the users while keeping the
response time for 90% requests less than 10s.
BASELINE: RHEL 7, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128GB DDR3-1333, Intel® SSD S3700 800GB, 1GB network speed.
NEW: RHEL 7, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128GB DDR4-1600, Intel® SSD S3700 800GB, 1GB network speed. Testing done by Neusoft 3/20/2015

86
TYDIC OCS* 2.1: OCS workload uses stressing tool to simulate typical business operations in OCS system. CPU utilization is around 90% and performance metric is TPS (Transactions per second)
BASELINE: RHEL 7, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128 GB DDR3-1333, Intel® SSD S3700 800GB, 10GB network speed
NEW: RHEL 7, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128 GB DDR4-2133, Intel® SSD S3700 800GB, 10GB network speed. Testing done by Intel 3/2/2015
Yonyou NC* v6.3: NC workload uses stress tool to simulate typical business operations in NC and the pressure focuses on application server side. CPU utilization is around 90% and performance metric is TPS
(Transactions per second).
NEW: RHEL 7, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 126 GB DDR4-2133, Intel® SSD S3700 800GB, 10GB network speed. Testing done by Intel 3/2/2015
Yonyou U9*: U9 workload uses stress tool to simulate typical business operations in U9 and the pressure focuses on application server side.
BASELINE: Yonyou U9 on Windows Server 2012 R2 (x64), 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128GB DDR3/1333, LVDIMM, SAS HD, 1Gb network
NEW: Yonyou U9 on Windows Server 2012 R2 (x64), 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128GB DDR4/1600, LVDIMM, SAS HD, 1Gb network. Source is Intel internal as of Aug 1, 2015.
TongTech TongWeb*: TongTech uses SPECjEnterprise2010 as benchmark workload. SPECjEnterprise2010 benchmark is a full system benchmark which allows performance measurement and characterization
of Java EE 5.0 servers and supporting infrastructure.
BASELINE: TongTech TongWeb on RHEL 7.0, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128GB DDR3/1333, LVDIMM, SAS HD, 1Gb network
NEW: TongTech TongWeb on RHEL 7.0, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128GB DDR4/1600, LVDIMM, SAS HD, 1Gb network. Source is Intel internal as of Aug 1, 2015.
TongTech TongIntegrator*: TongIntegrator workload simulates typical data integration scenarios, i.e. fetch data from source databasse, do necessary converstion as customized by end-user, store converted
data into destination database.
BASELINE: TongTech TongIntegrator on RHEL 7.0, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128GB DDR3/1333, LVDIMM, SAS HD, 1Gb network
NEW: TongTech TongIntegrator on RHEL 7.0, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128GB DDR4/1600, LVDIMM, SAS HD, 1Gb network. Source is Intel internal as of Aug 1, 2015.
Digital China Sm@rt* ESB: Digital China ESB workload uses its own stressing tool to simulate all key steps in information and process integration for different IT system. It processes the messages with
industrial standard format among different IT system destinations.
NEW: RHEL 7, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128 GB DDR4-1600, Intel® SSD S3700 800GB, 1GB network speed. Testing done by Digital China 3/18/2015
Huawei FusionSphere* 5.0: This workload uses self-developed stressing tool to simulate typical enterprise solution usage scenario. The workload is directly enforced request on backend FusionSphere solution
and record the success responses from the FusionSphere solution on the tested XEON server.
BASELINE: FusionSphere* 5.0, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128 GB DDR3-1333, SAS disk, 1GB network speed.
NEW: FusionSphere* 5.0, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128 GB DDR4-1600, SAS disk, 1GB network speed. Testing done by Huawei 3/25/2015
NARI Realtime Database*: NARI developed their stressing tool for performance testing. The tool simulates typical operations such as insert, query, update
BASELINE: NARI Realtime Database on RHEL 7.0, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128GB DDR3/1333, DIMM, Intel® SSD DC 3700, 10-Gigabit Intel® X540-AT2 network
NEW: NARI Realtime Database on RHEL 7.0, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128GB DDR4/1600, LVDIMM, Intel® SSD DC 3700, 10-Gigabit Intel® X540-AT2 network. Testing done by
NARI and Intel Aug. 13 2015
Neusoft HIS*: Neusoft developed a specific workload to simulate typical operations in HIS, i.e. patient registration, charge, statistics, etc.
BASELINE: Neusoft HIS on RHEL 7, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128GB DDR3/1333, DIMM, SAS HD, 1GB network
NEW: Neusoft HIS on RHEL 7, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128GB DDR4/1600, LVDIMM, SAS HD, 1GB network. Testing done by Neusoft and Intel Aug. 12 2015
SAP HANA*: SAP internal workload stressing HANA OLTP engine.
BASELINE: SLES 11 SP3, HANA SPS08, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512 GB regular DIMM-1600, 10GB network speed.
NEW: SLES 11 SP3, HANA SPS09, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 512 GB regular DIMM-1600, 10GB network speed. Testing done by Intel 1/8/2015

87
SAP SAE*: TPC-C workload
BASELINE: SAP ASE ASE 16.0 SP02, SLES11SP3, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512GB DDR3-1600, Intel® SSD DC P3700 , 10GB network speed
NEW: SAP ASE ASE 16.0 SP02, SLES11SP3, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 512GB DDR3-1600, ntel® SSD DC P3700 , 10GB network speed. Testing done by SAP Dec.19, 2014
Oracle MySQL*: Sysbench benchmark, Online Transaction Processing (OLTP) test profile. Defined 100 tables with 10 million rows each, which resulted in a database of 1 billion entries. This database was
260GB.
BASELINE: Ubuntu Linux* 14.10 x64, Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 256 GB DDR3-1333, 1GB network speed.
NEW: Ubuntu Linux* 14.10 x64, , Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 256 GB DDR3-1333, 1GB network speed. Testing done by Intel 4/24/2015
Altibase HDB-DA*: Workload for test is stock trade simulation, which consists of buying offer, buying and settlement processes from many concurrent users. The performance metric is the number of queries
per second that reflects multiple aspects of the capability of the system to process queries.
BASELINE: RHEL 6.4, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512 GB LVDIMM-1333, SSD, 1GB network speed.
NEW: RHEL 6.4, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 512 GB LVDIMM-16000, SSD, 1GB network speed. Testing done by Altibase 4/11/2015
Exasol EXASolution*: TPC-H like, 100GB load.
BASELINE: Exasol EXASolution on Ubuntu Linux 14.10 x64, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128GB DDR3/1333, regular DIMM, SAS HD, 1Gb network.
NEW: Exasol EXASolution on Ubuntu Linux 14.10 x64, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, GB DDR3/1333, regular DIMM, SAS HD, 1Gb network. Testing done by Exasol and Intel as of June
6, 2015.
Baosight iHyperDB*: Baosight developed their self-develped stressing tool iHyperDBSend for performance test. The tool will simulate typical operations such as insert, query, update.
BASELINE: Baosight iHyperDB on RHEL 7.0, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128GB DDR3/1333, LVDIMM, SAS HD, 1Gb network,
NEW: Baosight iHyperDB on RHEL 7.0, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128GB DDR4/1600, LVDIMM, SAS HD, 1Gb network. Source is Intel internal as of Aug 1, 2015.
IBM DB2 10.5 FP5: Proof of Performance and Scalability (POPS) is a data warehousing workload based on a true retail environment. The workload operates on a star schema database. A 10TB database (based
on raw data) was used for this test, and the results are an aggregation of 36 queries (10 fact-to-fact joins and 13 snowflake queries). All the queries were run in single user mode.
BASELINE: SLES 11.3, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 1 TB DDR3-1333, 1 SAS disk, 1GB network speed.
NEW: RHEL 6.5, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 1 TB DDR4-1600, 1 SAS disk, 1GB network speed. Testing done by IBM 3/20/2015
MariaDB* 10.1: Simulates typical R/O OLTP requests, 1M table size and 10M RW. Multi-table scenario.
BASELINE: Ubuntu Linux* 14.10 x64, Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 256 GB DDR3-1333, 1GB network speed.
NEW: Ubuntu Linux* 14.10 x64, , Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 256 GB DDR3-1333, 1GB network speed. Testing done by Intel 4/24/2015
AsiaInfo ADB*: The tool simulates typical OLTP operations such as insert, query, delete and update.
BASELINE: AsiaInfo ADB on RHEL 7.0, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 256GB DDR3/1333, DIMM, Intel® SSD DC 3700, 10-Gigabit Intel® X540-AT2 network
NEW: AsiaInfo ADB on RHEL 7.0, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 256GB DDR4/1600, LVDIMM, Intel® SSD DC 3700, 10-Gigabit Intel® X540-AT2 network. Testing done by AsiaInfo and
Intel Aug. 19 2015
Kingbase ES*: Kingbase developed their own performance test tool named TPCCLoader which simulate tpcc scenario to do the benchmarking.
BASELINE: Kingbase ES on RHEL 7, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128GB DDR3/1333, DIMM, SAS HD, 1GB network
NEW: Kingbase ES on RHEL 7, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128GB DDR4/1600, LVDIMM, SAS HD, 1GB network. Testing done by Kingbase and Intel Aug. 12 2015

88
Kingdom Technology STS*: KCBP Test is a performance test tool developed by Kingdom to test the performace of Kingdom STS specifically. The tool can launch numbers of threads to scale out processing
according to number of CPU cores of server
BASELINE: Kingdom Technology STS on Windows Server 2012 R2 (x64), 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 128GB DDR3/1333, DIMM, SAS HD, 1GB network
NEW: Kingdom Technology STS on Windows Server 2012 R2 (x64), 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 128GB DDR4/1600, LVDIMM, SAS HD, 1GB network. Testing done by Kingdom
Technology and Intel June 24 2015
Kx Systems kdb+* : The STAC-M3 Benchmark suite assesses the ability of a solution stack such as columnar database software, servers, and storage, to perform a variety of I/O-intensive and compute-intensive
operations on a large store of market data. The specifications are completely agnostic to architecture, which means that STAC-M3 can be used to compare different products or versions at any layer of the stack,
such as database software, processors, memory, hard disks, SSD, interconnects, and file systems.
BASELINE: Kx Systems kdb+* on CentOS Linux 6.5 2.6.32-504.el6.x86_64, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 1TB DDR3 1600 MHz 64x16GBDIMM, 7x 800GB S3700's, Raid 5 and 11
LUNs, 1Gb network
NEW: Kx Systems kdb+* on Red Hat Enterprise Linux 7.1 (kernel 3.10.0-229.el7.x86_64), 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores,1 TB DDR4 1600 MHz 64x16GBDIMM, 9 x 2tb P3700's, No Raid ,
JBOD, 9 LUNs, 1Gb network. Testing by STAC June 15, 2015
SunGard Adaptiv* 13.1: Adaptiv / BenchmarkX 13.1. Multiple Monte Carlo simulations each of 10K scenarios on 10K long vectors of data.
BASELINE: Windows Server 2012 Standard R2, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512 GB DDR3-1333, 2 SAS disks, 1GB network speed.
NEW: Windows Server 2012 Standard R2, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 384 GB DDR4-1600, 2 SAS disks, 1GB network speed. Testing done by Intel 2/12/2015
Vital Images VitreaAdvanced*: VitreaAdvanced is a 3D medical imaging software package comprising of approximately two dozen clinical applications, each involving 3D image processing algorithms and 3D
volumetric rendering.
BASELINE: VitreaAdvanced 6.6.331.1, Windows Server 2012 Standard R2, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512 GB DDR3-1600, 2 SAS disks, 1GB network speed.
NEW: VitreaAdvanced 6.7.34.2, Windows Server 2012 Standard R2, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 512 GB DDR4-2133, 2 SAS disks, 1GB network speed. Testing done by Vital Images
2/25/2015
Scripps Translational Science DNA Sequencing Pipeline*: Workload is a pipeline of bunch of software algorithms that help align, recalibrate the dna sequences to produce an output that could be used for
annotation and interpretation of genes.
BASELINE: RHEL 6.5, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 256 GB DDR3-1333, SAS disk, 1GB network speed
NEW: RHEL 6.5, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 1024 GB DDR4-1600, Intel SSD S3700, 1GB network speed. Testing done by Intel 3/2/2015
GROMACS*: Two representative public workloads are ion_channel and lignocellulose, both are part of European Unified Applications Benchmark Suite (http://www.prace-ri.eu/ueabs/#GROMACS) and were
represented at least in 2 CRT deals. The first one is 141K atoms computed with PME method, the second one is 3.1M atoms computed with RF method.
BASELINE: Red Hat 4.4.7-3, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512 GB DDR3-1333, 10GB network speed.
NEW: Red Hat 4.4.7-3, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 256 GB DDR4-1600, 10GB network speed. Testing done by Intel 2/15/2015
Epic* 2015: Workload: Transactional based Electronic Medical Records for Clinical systems. Benchmark is transactional and simulates a clinician's workflow. GREF is a "well- known" InterSystems* cache
benchmark for supported database transactions. GREF results provided by Epic is holistic system level sizing performance including all ancillary subsystems. Epic's framework for clinician workflow is solely
developed by Epic in .NET, HTML, native code for enhanced reliability and performance using InterSystem Cache for the database.
BASELINE: RHEL 7, VMware 5.5, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512 GB DDR3-1600, 50 SSDs, 10GB network speed.
NEW: RHEL 7, VMware 6.0, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 512 GB DDR4-2133, 50 SSDs, 10GB network speed. Testing done by Epic 3/2015

89
Dassault Systemes BIOVIA*: Bowtie 2 is an ultrafast and memory-efficient tool for aligning sequencing reads to long reference sequences. It is particularly good at aligning reads of about 50 up to 100s or
1,000s of characters to relatively long (e.g. mammalian) genomes. Bowtie 2 indexes the genome with an FM Index (based on the Burrows-Wheeler Transform or BWT) to keep its memory footprint small: for the
human genome, its memory footprint is typically around 3.2 gigabytes of RAM. Bowtie 2 supports gapped, local, and paired-end alignment modes. Multiple processors can be used simultaneously to achieve
greater alignment speed. Bowtie 2 outputs alignments in SAM format, enabling interoperation with a large number of other tools (e.g. SAMtools, GATK) that use SAM. Bowtie 2 is often the first step in pipelines
for comparative genomics, including for variation calling, ChIP-seq, RNA-seq, BS-seq, which is tightly integrated into some tools, including TopHat, Cufflinks, Crossbow, and Myrna.
BASELINE: RHEL 7, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512 GB DDR3-1066, Intel® SSD S3700, 1GB network speed.
NEW: RHEL 7, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 512 GB DDR4-1600, Intel® SSD S3700, 1GB network speed. Testing done by Intel 3/18/2015
Mentor Graphics Calibre* v2013.2_18.13: Design rule check.
BASELINE: Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512GB DDR3-1600, SAS disk, 1GB network speed.
NEW: Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 256GB DDR4-1600, 2 SAS disks, 1GB network speed. Testing done by Mentor Graphics 3/152015
ANSYS CFX* 16: FX 16 AirFoil, LeMansCar, Pump. Various CFD models.
BASELINE: RHEL 6.5 64 bit, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 512GB LVDIMM-1333, SATA disk, 1GB network speed.
NEW: RHEL 6.5 64 bit, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 256GB LVDIMM-1600, SATA disk, 10GB network speed. Testing done by Intel 2/25/2015
Altair RADIOSS* 13: NEON 1M 8ms DP.
BASELINE: RHEL 6.5 64 bit, 4S Intel® Xeon® processor E7-4890 v2, 2.8GHz, 15 cores, 264GB DDR3, SATA disk, 10GB network speed.
NEW: RHEL 6.5 64 bit, 4S Intel® Xeon® processor E7-8890 v3, 2.5GHz, 18 cores, 264GB DDR3, SATA disk, 10GB network speed. Testing done by Altair 4/8/2015

90
Legal Disclaimers: Performance
Performance tests and ratings are measured using specific
computer systems and/or components and reflect the approximate
performance of Intel products as measured by those tests. Any
difference in system hardware or software design or configuration
may affect actual performance. Buyers should consult other
sources of information to evaluate the performance of systems or
components they are considering purchasing. For more
information on performance tests and on the performance of Intel
products, Go to:
http://www.intel.com/performance/resources/benchmark_limitatio
ns.htm. and www.intel.com/software/products
Intel does not control or audit the design or implementation of
third party benchmarks or Web sites referenced in this document.
Intel encourages all of its customers to visit the referenced Web
sites or others where similar performance benchmarks are reported
and confirm whether the referenced benchmarks are accurate and
reflect performance of systems available for purchase.
Relative performance is calculated by assigning a baseline value of
1.0 to one benchmark result, and then dividing the actual
benchmark result for the baseline platform into each of the specific
benchmark results of each of the other platforms, and assigning
them a relative performance number that correlates with the
performance improvements reported.
The cost reduction scenarios described are intended to enable you
to get a better understanding of how the purchase of a given Intel
based product, combined with a number of situation-specific
variables, might affect future costs and savings. Circumstances will
vary and there may be unaccounted-for costs related to the use
and deployment of a given product. Nothing in this document
should be interpreted as either a promise of or contract for a given
level of costs or cost reduction.
Intel does not control or audit third-party benchmark data or the
web sites referenced in this document. You should visit the
referenced web site and confirm whether referenced data are
accurate.
INFORMATION IN THIS DOCUMENT IS PROVIDED “AS IS”. NO LICENSE,
EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL
PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. INTEL ASSUMES NO
LIABILITY WHATSOEVER AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED
WARRANTY, RELATING TO THIS INFORMATION INCLUDING LIABILITY OR
WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE,
MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR
OTHER INTELLECTUAL PROPERTY RIGHT.

91
All products, computer systems, dates, and figures specified are preliminary based on current expectations, and are subject to change without notice.
Intel, processors, chipsets, and desktop boards may contain design defects or errors known as errata, which may cause the product to deviate from published
specifications. Current characterized errata are available on request.
SPEC, SPECpower, and SPECjbb are trademarks of the Standard Performance Evaluation Corporation. See http://www.spec.org for more information.
TPC Benchmark is a trademark of the Transaction Processing Council. See http://www.tpc.org for more information.
Intel® Advanced Vector Extensions 2 (Intel® AVX2): Intel® Advanced Vector Extensions 2 (Intel® AVX2) provides higher throughput to certain processor operations. Due to
varying processor power characteristics, utilizing AVX instructions may cause a) some parts to operate at less than the rated frequency and b) some parts with Intel®
Turbo Boost Technology 2.0 to not achieve any or maximum turbo frequencies. Performance varies depending on hardware, software, and system configuration and you
can learn more at http://www.intel.com/go/turbo.
Intel® Hyper-Threading Technology (Intel® HT Technology): Available on select Intel® processors. Requires an Intel® HT Technology-enabled system. Your performance
varies depending on the specific hardware and software you use. Learn more by visiting http://www.intel.com/info/hyperthreading.
No license (express or implied, by estoppel or otherwise) to any intellectual property rights is granted by this document.
Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-
infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.
This document contains information on products, services and/or processes in development. All information provided here is subject to change without notice. Contact
your Intel representative to obtain the latest forecast, schedule, specifications and roadmaps.
The products and services described may contain defects or errors known as errata which may cause deviations from published specifications. Current characterized
errata are available on request.
Copies of documents which have an order number and are referenced in this document may be obtained by calling 1-800-548-4725 or by visiting
www.intel.com/design/literature.htm.
Intel, the Intel logo, Intel Xeon, the Intel Xeon logo are trademarks of Intel Corporation in the U.S. and/or other countries.
© 2015 Intel Corporation.
Legal Disclaimers

92
Optimization Notice
Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are
not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other
optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on
microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for
use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel
microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the
specific instruction sets covered by this notice.
Notice revision #20110804

Intel® Xeon® processor E7-8800/4800 v3 Application Showcase

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Intel® Xeon® processor E7-8800/4800 v3 Application Showcase (20)

More from Intel IT Center (19)

Recently uploaded (20)

Intel® Xeon® processor E7-8800/4800 v3 Application Showcase

Editor's Notes