IBM zEC12 zAware and Flash Express

IBM zAware and Flash Express

Mike Smith, System z Technical Specialist, September 2012

© 2012 IBM Corporation

IBM System z

Trademarks
The following are trademarks of the International Business Machines Corporation in the United States and/or other countries.

CICS* IBM* Parallel Sysplex* Tivoli*
DB2* IBM (logo)* PR/SM zEnterprise
Gdps* IMS RACF* z/OS*
HyperSwap Language Environment* System z* z/VM*

* Registered trademarks of IBM Corporation

The following are trademarks or registered trademarks of other companies.

Adobe, the Adobe logo, PostScript, and the PostScript logo are either registered trademarks or trademarks of Adobe Systems Incorporated in the United States, and/or other countries.
IT Infrastructure Library is a registered trademark of the Central Computer and Telecommunications Agency which is now part of the Office of Government Commerce.
Intel, Intel logo, Intel Inside, Intel Inside logo, Intel Centrino, Intel Centrino logo, Celeron, Intel Xeon, Intel SpeedStep, Itanium, and Pentium are trademarks or registered trademarks of
Intel Corporation or its subsidiaries in the United States and other countries.
Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both.
Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both.
Windows Server and the Windows logo are trademarks of the Microsoft group of countries.
ITIL is a registered trademark, and a registered community trademark of the Office of Government Commerce, and is registered in the U.S. Patent and Trademark Office.
UNIX is a registered trademark of The Open Group in the United States and other countries.
Java and all Java based trademarks and logos are trademarks or registered trademarks of Oracle and/or its affiliates.
Cell Broadband Engine is a trademark of Sony Computer Entertainment, Inc. in the United States, other countries, or both and is used under license therefrom.
Linear Tape-Open, LTO, the LTO Logo, Ultrium, and the Ultrium logo are trademarks of HP, IBM Corp. and Quantum in the U.S. and other countries.
* Other product and service names might be trademarks of IBM or other companies.
Notes:
Performance is in Internal Throughput Rate (ITR) ratio based on measurements and projections using standard IBM benchmarks in a controlled environment. The actual throughput that any
user will experience will vary depending upon considerations such as the amount of multiprogramming in the user's job stream, the I/O configuration, the storage configuration, and the
workload processed. Therefore, no assurance can be given that an individual user will achieve throughput improvements equivalent to the performance ratios stated here.
IBM hardware products are manufactured from new parts, or new and serviceable used parts. Regardless, our warranty terms apply.
All customer examples cited or described in this presentation are presented as illustrations of the manner in which some customers have used IBM products and the results they may have
achieved. Actual environmental costs and performance characteristics will vary depending on individual customer configurations and conditions.
This publication was produced in the United States. IBM may not offer the products, services or features discussed in this document in other countries, and the information may be subject
to change without notice. Consult your local IBM business contact for information on the product or services available in your area.
All statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only.
Information about non-IBM products is obtained from the manufacturers of those products or their published announcements. IBM has not tested those products and cannot confirm the
performance, compatibility, or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products.
Prices subject to change without notice. Contact your IBM representative or Business Partner for the most current pricing in your geography.

2 © 2012 IBM Corporation

IBM System z

Extending IBM System z Availability
• Companies are competing more aggressively to deliver
outstanding Service Levels
– Availability and performance are key to achieving service differentiation
– Market dynamics have changed to 24 x7, global environments, digital
markets
– Commoditization drives value differentiation in services which depend on
highly available systems
• According to Aberdeen Group*, business drivers behind most
companies’ high availability strategies are focused on reducing
unplanned downtime (80%), reducing financial risk of data loss
(75%) and the need for applications to be up and running 24/7
(65%)
• IBM has historically excelled at delivering high availability
– Systems Hardware
– Systems Software
– Operations
– The next step on the journey toward even higher availability
• Strengthen business application availability
• Simplify the problem determination process

IBM zAware and Flash Express help you further extend high
availability and performance to drive service levels to new heights
3 * Aberdeen Group Feb 2012 © 2012 IBM Corporation

IBM System z

Extending System z Availability with Flash Express and IBM zAware

Business
Server Systems Operations
Application
Highly resilient & available systems

Design Availability Availability
Availability

Sparing HyperSwap™ CoD Flash Express
Concurrent Upgrade CBU IBM zAware
Enhanced Book
Virtual networking GDPS®/HyperSwap
Availability
Data Sharing Call home
Hardware checks
Redundant parts
Parallel Sysplex®

Designed to Prevent Designed to Improve Designed to Improve Designed to Improve
Hard Failures System SW Availability Continuous Operations Business Availability

4 All statements regarding IBM’s future directions and intent are subject to change or withdrawal without notice and represent goals and objectives only © 2012 IBM Corporation

IBM System z

IBM Flash Express – Smarter Availability for Smarter Systems
Outstanding Availability and Performance - Innovative Flash Express

– Companies competing for the highest quality of service in today’s market must
deliver outstanding availability and performance
– Changes in workload processing can impact service levels at critical processing
times
• Flash Express is an innovative solution designed to help you
improve availability and performance to compete effectively in
24x7
24x7
today’s market
– Automatically improves availability for key workloads at critical processing times
– Drives availability and performance for workloads that cannot tolerate paging
spikes or inconsistent performance
– Slashes latency for critical application processing such as start of day processing
and also collection of diagnostics (SVC dumps, standalone dumps)
– Delivered as a new adapter card in the PCIe I/O drawer
• Benefits
– Improves availability and performance helping companies achieve highest
service levels
– Delivers a secured, resilient and immediately usable solution
– Automatic, requires minimal setup, no special training needed


IBM System z

Flash Express – What is it?
FLASH Express
• Physically comprised of internal storage on Flash SSDs
• Used to deliver a new tier of memory, storage class memory
• Uses standard PCIe I/O drawer
• Supported on z/OS® V1.13 plus web deliverable
• Flash Express cards delivered as a RAID 10 mirrored card pair
• Sized to accommodate all LPAR paging One Flash Express Card
– Each card pair provides 1.6 TB usable storage (3.2 TB total)
– Maximum 4 card pairs (4 X1.6=6.4 TB)
Access
• Immediately usable CPU
Time nanos
– No capacity planning needed
– No intelligent data placement needed Flash -
– Full virtualization of card across partitions microseconds
• Robust design
– Designed for long life Disk -milleseconds
– Designed for concurrent replacement or upgrade Flash memory blurs the distinction between
• Security Characteristics memory and storage characteristics

– Data encrypted on the flash express adapter with 128 bit AES encryption
– Keys stored on smart cards plugged into the System z SE
– Removal of smart cards renders data unusable


IBM System z

Flash Express Strengthens Availability
• Innovation to drive availability to exceptional levels
– Extends IBM’s expertise in memory management introducing a new
tier of memory using Flash Express
– Is an industry unique application of Flash to improve availability
– Takes the next step in advanced memory management
• Flash Express can improve availability and reduce latency
– Improves availability during transition periods and spikes
– Helps accelerate start of day processing - batch to online
– Enables faster snapshots of diagnostics (e.g. standalone dump)
– With pageable large pages can improve performance of DB2® and
Java™
– Ideal for applications with random read access or high read/write ratios
• Helps customers deliver vigorous service levels
– Designed to help provide exceptional availability
– Delivered with pageable large pages for superior performance
• Minimal configuration- no special skills needed
– Usable immediately
– Easy to set up and dynamically configurable


IBM System z

Performance Benefits from Application of Flash Express

• Achieve outstanding availability
– Flash memory is designed to improve availability and paging
performance
– Enough capacity so that all paging data can easily reside on a
pair of Flash Express cards
– CPU performance benefits expected from the use of pageable
large 1 MB pages
• Expected Benefits
– DB2 and Java:
• Estimated incremental several percent system CPU benefit
• Longer roadmap
• Continued optimization of 1MB pageable large pages vs.
4K pages
• Additional exploitation expected by middleware
• Available for ISV exploitation
• Designed for continued expanded use of new memory tier

IBM System z

Flash Express Exploitation
Flash support in z/OS sets the stage for further use

Flash Express z/OS
announced support

4Q12 1Q13
z/OS V1.13, enabling PTF
• IBM zEnterprise® EC12 z/OS RSM Enablement Offering
(zEC12) 9/19 • Flash Dynamic Reconfiguration
• z/OS support for zEC12 • Optional PLPA & Common dataset definition

z/OS V1.13 Flash Web Deliverable Expect continued middleware exploitation
December 14th 2012 for 1 MB pageable large pages

• Planned Flash Express and pageable large page
exploiters:
– DB2 for z/OS
– Java SDK7
– WAS Liberty Profile v8.5
– IMS™ 12
– z/OS V1.13 Language Environment®
– Other (CICS®)


IBM System z

Representative Use Cases - Flash Express

Flash Express can reduce latency delays from paging to bring
system availability to new heights and improve overall service levels

Application related errors will require collection of diagnostics.
These diagnostics can be collected faster with Flash Express,
reducing paging related delays that can impact your overall
availability.

Having your working data resident in Flash can help accelerate start
of day processing, and improve service for many industries at the
busiest time of their work day - a time when they cannot afford
disruptions.

DB2 and Java in memory buffer pools work to store and process
application data. DB2 and Java can benefit from 1MB pageable
large pages with Flash Express, improving overall CPU
performance.


IBM System z

Flash Express for the CIO

• CIOs are continually challenged to grow profitably in the face of increased competition in the
global marketplace. Facing a new economy and constrained resources they must innovate to
find new sources of value for their firms.
• Many CIOs believe innovation in service will drive that sustainable competitive advantage. By
consistently delivering service levels that exceed the expectations of customers, CIOs can
leverage new and enhanced services as the differentiator for their businesses.
• To meet service goals, CIOs must find a solution that quickly, non-disruptively and cost
effectively bolsters availability and performance.

Flash Express can automatically improve availability and performance enabling CIOs to
meet and exceed their commitments to deliver superior quality of service. Flash Express
is a compelling solution for improving availability that is secure, resilient, delivers rapid
time to value, and is usable right out of the box.


IBM System z

Flash Express for IT VPs

• You need to run IT as a service provider. You need availability and performance at maximum levels
regardless of application errors, shifts in processing, batch to first shift.
• You need the highest availability and performance capabilities so you can meet the critical IT service
level agreements your customers demand.
• You need to be highly available at first shift without any transitional delays typical from overnight
processing.
• You need to be able to balance work often even at the most critical times of the day while delivering
superior performance.
• Flash Express is a low risk, low cost solution and easily deployed solution to improving application
availability, performance and service levels.
• Resilient and secured, Flash Express is fully operational without requiring special skills for either
planning or operations.

For companies that require superior availability and performance, Flash Express is
uniquely designed to automatically strengthen availability and performance even
during periods that stress your system – such as system diagnostics, start of day
processing or other transitional periods.
All statements regarding IBM’s future directions and intent are subject to change or withdrawal without notice and represent goals and objectives only

IBM System z

IBM System z Advanced Workload Analysis Reporter (IBM zAware) -
Smarter Computing for Faster Service Restoration
• The complexity and rate of change of today’s IT infrastructures stress the limits of
IT to resolve problems quickly and accurately--while preserving SLAs
• IT is challenged to diagnose system anomalies and restore service quickly
– Systems may experience problems which are difficult or unusual to detect
– Existing tools do little to identify messages preceding system problems
– Some incidents begin with symptoms that remain undetected
– Manual log analysis is skills-intensive, and prone to errors
• IBM zAware with Expert System Diagnostics Gets it Right, Fast
• IBM zAware helps improve problem determination in near real time – helps rapidly and accurately
identify problems and speed time to recovery
• Analyzes massive amounts of data to identify problematic messages, providing information to
enable faster corrective action
• Analytics on log data provides a near real time view of current system state
• Cutting edge pattern recognition examines system behavior to help you pinpoint deviations
• Machine learning, modeling and historical data work to analyze your unique environment
• Benefits
– Can reduce problem determination and troubleshooting
– Particularly helpful when problems involve multiple teams
– Helps you diagnose problems quickly and more accurately to improve service recovery time
– Easy to use web based graphical interface


IBM System z

IBM zAware: An Expert integrated Analytics Solution
IBM System z Advanced Workload Analysis Reporter

• IBM zAware is a self learning, integrated expert solution that analyzes messages in
near real time to provide insight into the behavior of your system
– Analytic solution that adapts and learns your unique environment
– Host on zEnterprise EC12 server; can monitor other IBM System z® servers
• Runs on IFL or general purpose CP
• 4 GB memory base
– For up to 6 monitored z/OS systems Expert Machine
– Additional .25 GB per connected system after 6 Knowledge Learning

• 500 GB storage (estimated)
IBM
IBM
• Under 1% CPU overhead on monitored systems
zAware
zAware
• Self managed data store
– Shareable OSA ports required for communications Integrated Intuitive
– Dedicated IP address for partition
– Requires z/OS V1.13 + PTFs
– IE or Firefox browser


IBM System z

IBM zAware - Identify Unusual System Behavior
IBM zAware contains sophisticated analytics, applies IBM insight,
and machine learning to understand your unique system.

Monitoring Detection Frequency Reporting
• Supports IBM & non IBM • Detects anomalies other • Samples every 2 minutes • Near real time analysis
middleware and solutions might miss • 10 minute reporting • High level and drill down
applications • Can find the rare or interval intuitive web based
• Monitors OPERLOG in a infrequent message • Uses 90 day rolling reporting
sysplex or monoplex • Can detect an unusual baseline; utility provided to • Color coder, time slice
• Assigns a message number of normal populate baseline graphics
anomaly score to help messages • Allows dates to be • XML output can drive
you identify potential • Can detect messages included, excluded other ISVs or processes
issues issued out of context


IBM System z

Specific Applications of IBM zAware
• Identify a possible z/OS incident • Identify behavior after a change has
– Which image is having unusual behavior? been made
• Examines unique message behaviors – Are unusual messages being issued after
• High score generated by unusual messages or a change?
message patterns • New software levels (operating system,
– When did the behavior start? middleware, applications)
• For a selected 10 minute interval either the current • Updated system settings or system
10 minute interval or past intervals configurations
 Which messages are unusual? • Diagnose intermittent problems
 How often did the message occur?
• Are new unusual messages being issued when
 When did the message start to occur?
an intermittent problem occurs?
– Were similar messages issued
previously? – Are more messages issued then expected?
• Understands message characteristics and – Are messages issued out of a normal pattern?
message patterns

Vertical bar shows the Finds Anomalies that would be Hard to Detect
number of unique
messages in a 10 minute Trends Abnormalities Unusual situations
interval

Scoring of messages color
coded from common (blue)
to rare (orange)


IBM System z

IBM zAware Configuration for Maximum Flexibility

zEC12 IBM zAware host IBM zAware monitored client

z/OS z/OS z/OS z/OS z/OS

IBM
zAware
Host
Partition

IBM zAware
Web GUI to
monitor
results

Monitors zEC12 or any other System z servers running z/OS v1.13 +PTF
Support z/OS on VM
Requires OPERLOG


IBM System z

IBM zAware Sysplex View

Watch problems in near real
Bar height depicts number of time
unique message ids in that
interval

Follows typical Problem
Intuitive - drill down on Determination Workflow
interval (rectangle) of interest

Anomaly score indicates the rarity of the message IDs
within that interval
All statements regarding IBM’s future directions and intent are subject to change or withdrawal without notice and represent goals and objectives only
Screen captures are from development versions of software, displays may vary in production versions.

IBM System z

Representative Use Cases - IBM zAware

One member in the sysplex has communications related problems
causing symptoms (delays) showing up on another Sysplex
member. Problem was due to an error in the coupling facility exit

Online banking applications were timing out. Team could have
stopped subsystems to isolate problem. IBM zAware identified
LDAP as running short on memory, and this unusual memory
situation impacted other systems

Network slowdown problem evidenced itself as a probable TCP/IP
definition error. Cause was really security related, due to an
incorrectly coded RACF® access rule.

Typical problems identified are complex, cross sysplex and
sometimes are due to an unexpected cause. For instance many
slip trap messages might signal a problem but one or two might be
expected in certain conditions.


IBM System z

IBM zAware for the CIO

• CIOs are competing increasingly based on services in the global marketplace. Facing new pressures
to sustain high levels of availability and respond rapidly to IT service outages, you need to be able to
resolve system problems without painful delays.
• In a 24 X 7 environment, a system incident can drive up operations costs and disrupt service for hours
– even days. Resolution of complex system problems can incur high costs and impact reputation

• With 70–80%+ of IT expenditures spent on operations costs, and ever increasing IT complexity, CIOs
are facing fiscal pressures to reduce spend, yet meet demanding service levels. IBM zAware is
designed to support this growing sophistication of IT while reducing operations complexity.
• IBM zAware can help you quickly restore service levels by applying integrated expert analytics
in near real time to quickly and accurately isolate system anomalies. Using IBM zAware you can
take action to prevent problem recurrence and optimize service delivery.

IBM zAware is an integrated, self learning, analytics solution that helps identify unusual
system behavior to help improve service levels. IBM zAware uses machine learning to
help your organization gain visibility into system behavior helping you optimize service.
*IDC 2009


IBM System z

IBM zAware for IT VPs and
staff

• You need to improve IT availability to compete as a service provider. You need to be able to rapidly
identify any system problems and reduce the time to diagnose problems. You need to address
problems quickly and accurately, to keep availability at its highest levels.
• You function as a service provider whether your customers are internal or external. You cannot afford
service disruptions that take too long to diagnose and repair.
• Yet unsolved system outages and performance degradation can limit your ability to meet critical IT
service level agreements and remain competitive.
• You need to monitor system health with vigilance to understand typical system behavior and identify
possible deviations.
• You must identify problems and quickly restore service even when symptoms are sporadic or demand
high skills.
• IBM zAware is an integrated expert solution that uses analytics to help you identify potential
problems helping you to improve overall service levels

For companies that require superior service levels and fast problem resolution, IBM zAware uses
expert analytics to learn your system characteristics to help you identify problematic system
behavior. IBM zAware can help you detect problematic trends and resolve issues quickly so you
can restore service levels without delay.

IBM System z

IBM zAware Can Reduce Time to Repair to Improve Availability

Without IBM zAware With IBM zAware

Problem Problem

Decide Business Application Decide Business Application
what to do Impact Steady State what to do Impact Steady State

Little Often, More
Detect
advanced Multiple Precise
changing
warning Attempts to Corrective
conditions
Correct Action
Problem

restart
shutdown shutdown restart
gathering of
gathering of
Time diagnostic
Time
diagnostic
information information

Ineffective time spent in problem More precise and early diagnosis can
determination and trial and error. shorten impact time and help you to
Incorrect problem identification may avoid a similar problem. Detect trends
result in the wrong fixes being applied. and problems as they are unfolding.


IBM System z

Summary: Why Flash Express and IBM zAware are Important

Flash Express
• Today’s market transformation, evolution to cloud and new service delivery
models will demand increased availability
• Pressures from social networking and mobile computing applications drive the
demand for high performance and highly available applications
• Big Data and other analytic applications will require random access of data, and
improved paging performance, leveraging the use of pageable large pages
(PLP) and Flash Express

IBM zAware
• As companies innovate, new interaction and complexity between applications
can result in complex problems that need to be resolved quicker and more
accurately
• Global IT departments have a reduced likelihood of having the right experts
available at the same time to resolve complex problems with many
interdependencies
• Organizations need a solution that is self tuning, with expert analytics to
diagnose anomalies. The complexity and flux of today’s problems makes them
virtually impossible to codify in advance


IBM System z

IBM z/OS Solution Family Can Improve Problem Diagnostics
Rules Analytics/ Examines Self Method
Solutions Available: based Statistical message Learning
model traffic
Included in z/OS • Checks configurations Screens for
- z/OS Health Checker • Programmatic, applies to IBM and ISV tools  conditions
• Can escalate notifications

- z/OS PFA • Trending analysis of z/OS system resources, Early detection of
and performance   a problem
• Checks metrics like Storage, Frames
• Can invoke z/OS RTD

- z/OS RTD • Real time diagnostics of specific z/OS system Examines the
issues  * operating
system when
an incident is
reported

- IBM zAware • Pattern based message analysis Diagnose an
• Self learning    incident
• Provides aid in diagnosing complex z/OS anywhere in
problems, including cross sysplex, “sick but the z/OS
not dead” software stack
while it is
occurring

IBM zAware uniquely analyzes messages in context to determine unusual behaviors
IBM zAware uniquely understands and tunes its own baseline to compare against current activity
IBM zAware does not depend on other solutions, coding of rules, and is always enabled to watch your system

* Limited

IBM System z

Cost of Downtime

Reduce availability lapses by 45 minutes and pay for Flash Express
Datacenter Downtime How
Much Does IT Really Cost?

Average one hour of
downtime per event at
$181,770 per hour with two
more hours to restore.

Total cost over $.5M


IBM System z


IBM System z

Flash Express
Technical Background


IBM System z

Introducing Flash Express
Time to Read
 Flash Express is intended to improve System z availability Data measured
– Slash latency delays from paging in System z
– Flash Memory is much faster than HDD Instructions
– Flash Memory is much slower than main memory
– Flash Memory takes less power than either
– Make your start of day processing fast Real Memory:
– Eliminate frustrating delays from SVC Dump processing

More Latency
(256B line)
 zEC12 offers optional Flash Express memory cards ~ 100
– Supported in PCIe I/O drawer with other PCIe I/O cards Instructions
– Pairs of cards for availability
Flash Memory
– No HCD/IOCP definitions required
(4K page)
 Assign Flash Memory to partitions like main memory
– Assignment is by memory amount, not by feature ~ 100K
– Each partition’s Flash Memory is isolated like main memory Instructions
– Dynamically increase the partition maximum amount of Flash External Disk
– Dynamically configure Flash Memory into and out of the partition
(4K page)
~ 5,000K
Instructions

28

IBM System z

What Is Flash Express?

• Flash Express is internal storage implemented via NAND Flash SSDs
(Solid State Drives)
– Flash Cards plug as pairs into IBM’s PCIe I/O drawers
– A pair provides 1.6 TB of mirrored storage
– A maximum of 4 pairs are supported in a zEC12
• Flash is accessed using the new System z architected EADM (Extended
Asynchronous Data Mover) Facility
– An extension of a proven architecture used with expanded storage
– Access is initiated with a Start Subchannel instruction.
• The main application of Flash is as a paging store for z/OS
– Flash Express provides advantages in resiliency and speed
– With large pages being introduced in tandem for exceptional performance


IBM System z

Flash Allows Flexibility in Your Configurations

Allocating Flash to a partition
 The initial and maximum amount of Flash
Memory available to a logical partition is
specified at the SE or HMC
 Can also dynamically change maximum
amount of Flash Memory available to a
specific logical partition
 Additional Flash Memory (up to the
maximum allowed ) can be configured
online to a logical partition dynamically at
the SE or HMC
 Can dynamically configure Flash Memory
offline to a logical partition at the SE or
HMC
 Uses predefined subchannels, no IOCDS
Manage Flash Allocation


IBM System z

Flash: A z/OS Virtualized Resource

• There is full virtualization of physical Storage Class
Flash PCIe cards across partitions Memory Storage Class
Memory
• Each logical partition can be configured SCM z/OS
with its own address space for Storage SPACE z/OS SCM
SPACE
class memory (Flash memory)
• Flash is allocated to partitions by
Main Main
amount of memory, not by card size Memory Memory
• Architecture supports an ability to Partition
evolve the underlying Flash technology Maximum
while preserving the API
Partition partition1 partition2
• No Hardware Specific design elements Maximum
are in software zEC12
– Error Isolation, Transparent
mirroring, Centralized diagnostics,
– Hardware Logging, FRU Call, Partition
Recovery are Independent of Partitio Initial
n Initial Value
software Value
Partition
Initial Data transfer via EADMF
Value


IBM System z

Smart Flash vs Disk Placement Criteria

Intelligent placement of pages to Flash Express and/or Disk
Data Type Data Page Placement
PLPA PLPA pages will be placed both on flash and disk.

VIO VIO data will always be placed on disk

Pageable Large Pages If contiguous space is available, pageable large
pages will be written to Flash Express.

If Flash Express is not available in the system
configuration, then pageable large pages are backed
with 4k page frames.
Other decision points are If available space exists on both Flash Express and
automated disk then the paging decision selection is made based
on response time.


IBM System z

Flash Express Exploitation on zEC12 IBM Flash Express
Sets the stage for further exploitation

 Flash Express will be exploited by z/OS
 z/OS V1.13 Flash Web Deliverable – GA December 14, 2012
 Pageable Large Pages (1 MB)
 z/OS V1.13 enabling PTFs for RSM enhancements – 1Q2013
 Flash Dynamic Reconfiguration
 Optional PLPA and COMMON Page data sets
 2GB Large Page Support (Targeted 1Q2013)

 DB2 for z/OS V10 and JAVA SDK7 SR3 will support pageable Large Pages (SoD)
 IBM is working with it’s Linux Distribution partners to include support in future Linux on
System z distribution releases
 IMS* CQS will use pageable Large Pages when IMS runs on zEC12 –
Availability targeted for end of 2013

*All statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only.

33

IBM System z

zAware
Technical Background


IBM System z

IBM zAware – Smarter Computing Needs Smart Monitoring

• New technology based on patented machine
learning developed by IBM Research
• Cutting edge pattern recognition techniques look
at the health of a system to pinpoint deviations
from the ‘norm’
• High speed analytics applied to IT facilitates the
ability to consume and reduce large quantities
of messages
• Pinpoints potentially problematic messages to
speed up the problem determination process
and improve time to restore service.
• Can help identify unusual or problematic
messages early before they cascade into larger IBM zAware can watch your
problems system to detect unusual
behavior of z/OS images in
• Helps IT personnel take corrective actions near real time
swiftly


IBM System z

IBM zAware –
IBM System z Advanced Workload Analysis Reporter

• Monitors z/OS OPERLOG including all
messages written to z/OS console, including
ISV and application generated messages
• Detects things typical monitoring systems
miss due to:
– Too many normal messages
– Messages issued outside of the normal
pattern of occurrence
• Uniqueness (message not common
enough)
– Useful for real-time event diagnostics
• Color-coded, easy-to-use web browser GUI
• XML output can feed other products
Ability to drill down for details on
anomalies.


IBM System z

Sample Output - Interval View
When drill down, see JES2 resource shortage, and the messages indicating what caused it.

Detailed view
List of all messages issued during
interval
When messages occurred
Number of appearances
Anomaly score indicates rarity of
message
Contribution to interval score
Rule indicate IBM knowledge applied

Hotlink to IBM message text

Message context -
accompanying messages
Most likely messages listed
37
at beginning © 2012 IBM Corporation

IBM System z

IBM zAware Complements Your Existing Environment
Enterprise Management
Workspace views
Service Management Software

IBM zAware provides XML data through an
API which can feed Tivoli® or other
software products or processes

zEC12
z/OSMF

zLinux
zLinux
zLinux
IBM zAware Machine

z/OS
z/OS
z/OS
Virtual Machine

Virtual Machine
Sysplex Management

Virtual zOS
zOS
zOS
ICF
Sysplex views
Management of z/OS z/VM®
PR/SM
System z CPU, Memory and IO and IO
Z CPU,
Memory
SE
IBM zAware

zLinux

zLinux
zLinux
Machine

z/OS
z/OS
z/OS

z/OS
Virtual Machine
Virtual zOS
zOS
ICF
Customer controlled scope z/VM
Sysplex views PR/SM™
z/OS Image view z CPU, ZMemoryand IO IO
CPU, Memory and

SE


IBM System z

A Closer Look Inside IBM zAware
View Control IBM
IBM zAware zAware-specific
results knobs via IBM
zAware GUI IBM zAware uses it’s own LPAR
z/OSMF
IBM zAware GUI
Self contained firmware stack

zEC12 zEC12/z196/z114
rk
e two
rn IBM zAware
z/OS z/OS z/OS z/OS z/OS
me Server
sto
Cu Web Server

Operlog Operlog Operlog Operlog Operlog
Analytics
Results
Persistent
Storage Models
File LOGGER LOGGER LOGGER LOGGER LOGGER
Data Data Data Data Data Data
System Retrieval Transport Transport Transport Transport Transport

LPAR LPAR
el
unn HiperSockets HiperSockets
I PT
OSA (for data from other OSA (for data from other
servers) servers)

Manage IBM zAware
Firmware partition z/OS 1.13 plus PTFs or higher for monitored clients
(similar to CF)

39

IBM System z

IBM zAware Server/Client Overview

IBM zAware
Server
Firmware z/OS Client z/OS Client
Analytics
Models Operlog Operlog

Actions
LOGGER LOGGER
Data
Retrieval Data Transport Data Transport

PR/SM
OSA, HiperSocket

Web Server
File System
Alert Console

40

IBM System z

IBM zAware Infrastructure

Firmware Stack z/OS Client IXGLOGR @space
Consoles @space
IXGWRITE Thread Operlog data
IBM zAware Put data in logstream
Server copy Operlog records
IXGWRITE

put on queue

Logger Task
ZAI SERVER(host-location)

Analytics dub SYSPLEX.OPERLOG
loop Log stream
Al

Engine
er
t

awakened
Client
Server ZAI(YES)
Parser USS ZAIDATA(‘OPERLOG’)
Transport
Transport

PR/SM
Network Transport
41

IBM System z

IBM zAware Operating Requirements
• zEC12 to host IBM zAware Server
– IBM zAware requires it’s own LPAR and runs it’s own self-contained firmware stack.
• This will reduce the number of LPARs available for customer use
– IBM zAware processor resources can be IFL or General Purpose CP
– Memory and DASD resources are dependent on the number of monitored clients, amount
of message traffic, length of time data retained
• Memory - Min 6GB + 256 MB
• DASD ~ 500GB (ECKD or FC)
• IBM zAware uses Logical Volume Manager (LVM) to aggregate multiple physical devices into a
single logical device
– Network: HiperSockets or OSA ports – for both gathering of instrumentation data, and
outbound alerting/communications
• Need dedicated IP address for partition
• zAware Monitored Clients
 IBM zAware monitored clients can be on any System z Server running z/OS 1.13 + PTFs
 IBM zEnterprise 196 (z196), IBM zEnterprise 114 (z114), etc., and can share log files via IP network with IBM
zAware server

• 90 days historical syslog or OPERLOG data to initially prime IBM
zAware
42

IBM zEC12 zAware and Flash Express

More Related Content

What's hot (16)

Similar to IBM zEC12 zAware and Flash Express (20)

Recently uploaded (20)

IBM zEC12 zAware and Flash Express