Migrating Mission-Critical Workloads to Intel Architecture
1 like1,093 views
The document outlines a guide for migrating mission-critical workloads to Intel architecture, emphasizing the importance of planning for high-availability systems. It highlights the performance advantages of Intel Xeon processors over legacy RISC/Unix systems and presents an eight-step framework for building the business case and assessing migration feasibility. Additionally, it details six steps for the actual migration process, including assessing server load, developing a proof of concept, and executing the migration to ensure successful deployment.
Migrating Mission-Critical Workloads to Intel Architecture
- 1. Migrating Mission-Critical Workloads to Intel® Architecture PLANNING THE MOVE TO MODERN HIGH-AVAILABILITY SYSTEMS
- 2. 1 No computer system can provide absolute reliability, availability, or serviceability. Requires an Intel® Run Sure Technology–enabled system, including an enabled Intel processor and enabled technology(ies). Built-in reliability features available on select Intel processors may require additional software, hardware, services, and/or an Internet connection. Results may vary depending upon configuration. Consult your system manufacturer for more details. 2 “Up to 1.8x performance leadership” claim based on the SPECint* 2006 rate base benchmark for the IBM* Power* 750 Express* server with POWER7+* processors and an estimate for four-socket server based on Intel Xeon® processor E7-4890 v2. Publication for SPECint_rate_base2006 for Power 750 server with IBM POWER7+ processors (1,230 baseline score) source: spec.org/cpu2006/results/res2013q3/cpu2006-20130805-26129.html. Intel estimate based on SPECint_rate_base2006 for Intel Xeon processor E7-4890 v2 (2,238 baseline score estimate). 3 “Up to 1.28x performance leadership” claim based on SPECint2006 rate base benchmark estimate for the Oracle* SPARC* T5-4 server with SPARC T5 processors and an estimate for four-socket server based on Intel Xeon processor E7-4890 v2. Intel estimate based on SPECint_rate_base2006 for SPARC T5-4 server with SPARC T5 processors (1,745 baseline score estimate). Intel estimate based on SPECint_rate_base2006 for Intel Xeon processor E7-4890 v2 (2,238 baseline score estimate). 2 Performance Disclaimers
- 3. The Data Center under Pressure Cloud and big data Mission-critical workloads NEW SERVICES DEMAND RELIABLE BUSINESS SYSTEMS Reduced operational costs TIGHTER BUDGETS 3
- 4. Inefficient Legacy Infrastructure • High availability • Reliability 4 RISC/UNIX* SYSTEMS FOR MISSION-CRITICAL WORKLOADS PROS: CONS: • Expensive • Proprietary • Difficult to maintain • Lack flexibility and scalability
- 5. Innovations for x86 Architecture PERFORMANCE 5 COMMON PLATFORM BASED ON OPEN INDUSTRY STANDARDS SMALLER DATA CENTER FOOTPRINT MORE FLEXIBLE IT ENVIRONMENT VIRTUALIZATION CAPABILITIES RELIABILITY, AVAILABILITY, AND SERVICEABILITY (RAS) IMPROVEMENTS
- 6. Intel® Xeon® Processor-Based Solutions High-availability RAS capabilities Lower total cost of ownership (TCO) than legacy RISC/UNIX* systems Enabling new enhancements in operating systems 6
- 7. Four-Socket Server Performance 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. The performance levels for Intel® Xeon® processor-based servers compared to IBM* Power* and Oracle* SPARC* servers 7
- 8. The Intel® Xeon® Processor E7 v2 Family Intel® Xeon® processor E7 v2 family-based servers are an ideal platform for enterprise mission-critical solutions. • World-class uptime and data integrity for critical business workloads • Robust in-memory performance for real-time business intelligence and transaction-intensive workloads • High performance and scalability for cloud environments • Accelerated data encryption and platform hardening • Fast return on investment (ROI) and low TCO 8
- 9. Building the Business Case: Eight Steps Understand and align business and technical goals. • Refresh aging infrastructure? • Improve performance? • Reduce space? • Standardize or virtualize your infrastructure? • Get the latest versions of database, operating system, or application software to take advantage of new mission-critical capabilities? Define scope. • Keep it simple, or tackle as a series of projects. Define success criteria. • Quantify measurable values. (TCO by a specific percentage? Transactions per second?) STEP 1: 9 STEP 2: STEP 3:
- 10. Building the Business Case: Eight Steps (Continued) Quantify and track ROI and TCO. • Develop models to evaluate potential ROI and TCO. Identify and engage stakeholders. • Collaborate early. • Understand business pain points and opportunities. • Get help defining objectives. • Establish project and budget support. Assess your existing environment and workload. • Determine requirements. • Evaluate current against future environment. STEP 4: STEP 5: STEP 6: 10
- 11. Determine feasibility and risk. • Can you meet requirements? • Do you have the skill sets needed? Decide how to proceed. • Revisit TCO and ROI. • Weigh gains against risks. 11 Building the Business Case: Eight Steps (Continued) STEP 7: STEP 8:
- 12. Migration to x86 Architecture: Six Steps Assess current server load. • Determine preliminary server sizing, memory, and disk space. • Take advantage of sizing tools such as the Solaris* SYSSTAT package. Identify pilot opportunities. • Look for an early win. • Evaluate strategic importance and user readiness. Develop a proof of concept (PoC). • Test application on the target hardware. • Ensure business unit testing. • Finalize server sizing. 12 STEP 1: STEP 2: STEP 3:
- 13. Evaluate full solution architecture needs. • Robust hardware and redundancy • Clustering software (Linux* operating system option) for redundancy • Horizontal and vertical scaling Rehearse your migration. • Streamline and improve on results of PoC. Cut over to production. • Tighten project plan. • Notify business units. • Execute migration. • Conduct QA and acceptance. • Cut over or run in parallel until sign-off. 13 Migration to x86 Architecture: Six Steps (Continued) STEP 4: STEP 5: STEP 6:
- 14. Migrating Mission-Critical Deployments to x86 Architecture Planning Guide Read the complete Planning Guide: Migrating Mission-Critical Solutions to x86 Architecture at intel.com/riscunixguide. EVERYTHING YOU NEED TO GET STARTED intel.com/ITCenter 14
- 15. Legal SPEC, SPECint, SPECfp, SPECrate, SPECpower, SPECjAppServer, SPECjEnterprise, SPECjbb, SPECompM, SPECompL, and SPEC MPI are trademarks of the Standard Performance Evaluation Corporation. See spec.org for more information. This presentation is for informational purposes only. THIS DOCUMENT IS PROVIDED “AS IS” WITH NO WARRANTIES WHATSOEVER, INCLUDING ANY WARRANTY OF MERCHANTABILITY, NONINFRINGEMENT, FITNESS FOR ANY PARTICULAR PURPOSE, OR ANY WARRANTY OTHERWISE ARISING OUT OF ANY PROPOSAL, SPECIFICATION, OR SAMPLE. Intel disclaims all liability, including liability for infringement of any property rights, relating to use of this information. No license, express or implied, by estoppel or otherwise, to any intellectual property rights is granted herein. Copyright © 2014 Intel Corporation. Intel, the Intel logo, the Intel Inside logo, the Look Inside. logo, and Xeon are trademarks of Intel Corporation in the U.S. and/or other countries. *Other names and brands may be claimed as the property of others. 15