Intel Xeon Processor Update

Intel Xeon Processor Update Gareth Tucker Technical Account Manager IBM Europe, Middle East & Africa 1

This slide MUST be used with any slides removed from this presentation Legal Disclaimer Intel may make changes to specifications and product descriptions at any time, without notice. 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, visit Intel Performance Benchmark Limitations 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. Intel processor numbers are not a measure of performance. Processor numbers differentiate features within each processor family, not across different processor families. See www.intel.com/products/processor_number for details. 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. Hyper-Threading Technology requires a computer system with a processor supporting HT Technology and an HT Technology-enabled chipset, BIOS and operating system. Performance will vary depending on the specific hardware and software you use. For more information including details on which processors support HT Technology, see here Intel Turbo Boost Technology requires a PC with a processor with Intel Turbo Boost Technology capability. Intel Turbo Boost Technology performance varies depending on hardware, software and overall system configuration. Check with your PC manufacturer on whether your system delivers Intel Turbo Boost Technology. For more information, see http://www.intel.com/technology/turboboost Intel Virtualization Technology requires a computer system with a processor, chipset, BIOS, virtual machine monitor (VMM) and applications enabled for virtualization technology. Functionality, performance or other virtualization technology benefits will vary depending on hardware and software configurations. Virtualization technology-enabled BIOS and VMM applications are currently in development. 64-bit computing on Intel architecture requires a computer system with a processor, chipset, BIOS, operating system, device drivers and applications enabled for Intel 64 architecture. Performance will vary depending on your hardware and software configurations. Consult with your system vendor for more information. Lead-free: 45nm product is manufactured on a lead-free process. Lead is below 1000 PPM per EU RoHS directive (2002/95/EC, Annex A). Some EU RoHS exemptions for lead may apply to other components used in the product package. Halogen-free: Applies only to halogenated flame retardants and PVC in components. Halogens are below 900 PPM bromine and 900 PPM chlorine. Intel, Intel Xeon, Intel Core microarchitecture, and the Intel logo are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. 2009 Standard Performance Evaluation Corporation (SPEC) logo is reprinted with permission 2

Intel Xeon 5600 Energy Efficiency Building on Xeon 5500 Leadership Capabilities Lower Power CPUs Better performance/watt Lower power consumption 130W 95W 80W 60W (6C) 40W (4C) Intelligent Power Technology Integrated Power Gates and Automated Low Power States with Six Cores Intel Xeon 5600 Intel Xeon 5600 CPU Power Management Reduced power consumption through more efficient Turbo Boost and memory power management Lower Power DDR3 Memory Up to 1.5W per DIMM reduction in memory power 1 Intel Xeon 5600 delivers greater platform Energy Efficiency 3 Lower power CPU TDP options for Xeon 5600 1 DDR3L supported for Xeon 5600 only. System level power testing sing Samsung 1.35V DIMMs as compared to Samsung 1.5V DIMMs reduced power by 0.52W per DIMM at active idle, and 1.42W per DIMM under 100% load. Source: Intel internal measurements Feb 2010 using server side java benchmark across a load line. Power measurement at the wall using same system configuration; memory was the only variable changed. See backup for system configuration. LAN and Storage Capabilities Same format Alternate format

New Security Features with Intel Xeon Processor 5600 Advanced Encryption Standard New Instructions (AES-NI) Intel Trusted Execution Technology (TXT) Enables broad usage of encryption throughout the enterprise Prevents the insertion of malicious software prior to VMM launch Ready for Today Prepared for Tomorrow 4

Intel AES-NI Enabling Encryption Throughout the Enterprise TODAY With the performance penalty, I can t afford to use encryption Xeon 5600 Intel AES-NI helps removes the barriers for protecting data" secure transactions throughout my environment support more users than Xeon 5500, even using SSL transactions 1 application encryption (e.g. database) take advantage of database encryption features 2 full disk encryption shorter hard drive provisioning for increased data protection 3 AES-NI Enables Encryption as the Default Value 1 Intel Xeon X5680 can support 23% more users than Intel Xeon X5570, even using SSL transactions. Source: Internal Intel measurements with a web banking workload, comparing Intel Xeon Xeon X5680 (3.33 GHz) with SSL ON with Intel Xeon X5570 with SSL OFF. 2 Up to 89% reduction in decryption performance using Oracle 11g with TDE, comparing the time to decrypt a 5.1 million row encrypted table with AES-256 CBC mode on Intel Xeon Xeon X5680 (3.33 GHz) with Intel Performance Primitives crypto library (IPP), and Intel Xeon X5570 (2.93 GHz) without IPP. 3 Full Disk Encryption: Up to 42% faster server hard drive provisioning time for encrypting a 32GB X25E SSD using an Intel Xeon Xeon X5680 (3.33 GHz) compared to an Intel Xeon X5570. 5 See backup foil for details on all configurations

Number of Users Decryption Time(us) provisioning time (seconds) Intel Xeon Processor 5600 Series Intel AES-NI Encryption Performance Web Banking Workload (MS IIS/PHP) 1 4.5x 23% Higher is better Database Encryption/Decryption (Oracle 11g) 2-89% Lower is better Full Disk Encryption (McAfee Endpoint Encryption) 3-42% Lower is better 13000 16000 158 17 3500 20 12 WDC w/o encryption NHM w/o encryption WSM NHM w/o Intel IPP WSM NHM WSM 1 System configuration: Windows 2008 R2 x64 Ent. Server. PHP banking sessions /users measured with Intel Xeon X5680 (3.33 GHz) vs Intel Xeon 5160 (3.00 GHz) and Intel Xeon X5570 (2.93 GHz), 24 SSD RAID 0 arrays, TLS_RSA_with_AES_128_CBC_SHA cipher suite. 2 System configuration: Oracle 11g with TDE, time takes to decrypt a 5.1 million row encrypted table with AES-256 CBC mode on WSM 3.33 GHz optimized with Intel Performance Primitives crypto library (IPP) vs NHM 2.8 GHz without IPP. Timing measured is per 4K of data. 3 System configuration: McAfee Endpoint Encryption for PCs (EEPC) 6.0 package with McAfee epolicy Orchestrator (epo) 4.5 encrypting a 32GB X25E SSD with WSM 3.33 GHz vs. NHM 2.93 GHz. 24GB of memory. 6

Refresh and Future-Proof with Intel Xeon Processor 5600 Series 2007 2008 2009 2010 Xeon 5300 Xeon 5400 Xeon 5500 Xeon 5600 REFRESH RETIRE Intel VT FlexMigration Greater VM capacity / performance, lower power Intel VT FlexMigration compatibility Advanced capabilities (Intel VT, Intel AES-NI, Intel TXT) Refresh Your Pools to Support New Services and Business Growth Source: Intel measurements as of Feb 2010. Performance comparison using server side java bops (business operations per second). Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. For detailed calculations, configurations and assumptions refer to the legal information slide in backup. 7

Xeon 5500 Xeon 5600 SKU Transition Xeon 5600 Usage Advanced 6.4 GT/s QPI 8MB / 12MB DDR3 1333 Turbo Boost HT Standard 5.86 GT/s QPI 8MB / 12MB DDR3 1066 Turbo Boost HT Xeon 5500 X5570 2.93 GHz X5560 2.80 GHz X5550 2.66 GHz E5540 2.53 GHz E5530 2.40 GHz E5520 2.26 GHz 95W 95W 95W 80W 80W 80W Maximum Performance 6 Cores 4 Cores (freq optimized) Best Price Performance Higher Freq More Cache X5680 6C 3.33 GHz X5670 6C 2.93 GHz X5660 6C 2.80 GHz X5650 6C 2.66 GHz E5640 4C 2.66 GHz E5630 4C 2.53 GHz E5620 4C 2.40 GHz 130W 95W 95W 95W 80W 80W 80W 80W X5677 4C 3.46 GHz X5667 4C 3.06 GHz L5640 6C 2.26 GHz 1 130W 95W 60W L5630 4C 2.13 GHz 2 40W L5609 4C 1.86 GHz 3 40W Freq-Optimized Low Power Options Basic 4.8 GT/s QPI 4M cache DDR3 800 E5506 2.13 GHz E5504 2.00 GHz 80W 80W E5502 1.86 (2C) 80W Cost-Optimized Higher Frequency Xeon E5500 SKUs E5507 2.26 GHz (4C) E5506 2.13GHz (4C) 80W 80W E5503 2.00 GHz (2C) 80W 8 Xeon 5600 (Westmere-EP) SKUs Xeon 5500 SKUs 1 SKU supports 5.86 GT/s QPI speed and DDR3 1333MHz 2 SKU supports 5.86 GT/s QPI speed and DDR3 1066MHz 3 SKU supports 4.80 GT/s QPI speed and DDR3 1066MHz, with Turbo and HT off.

Intel Xeon Processor 5600 Series Performance Summary Technical Compute Servers Mainstream Enterprise Servers 1.00 Up to 63% Performance Memory Bandwidth CAD Floating Point 1.20 1.21 1.25 Life Sciences 1.44 HPC Financial 1.61 1.63 1.00 Up to 46% Performance ERP 1.27 Integer Virtualization Energy Efficiency Java 1.40 1.42 1.42 1.46 Bandwidth Sensitive Frequency Sensitive Xeon 5600-EP (6C, 3.33 GHz) vs. Intel Xeon X5570 (4C, 2.93 GHz) Xeon 5600-EP (6C, 3.33 GHz) vs. Intel Xeon X5570 (4C, 2.93 GHz) Baseline Stream Computer Aided Eng. SPECfp rate 2006 Life LINPACK Black Sciences Scholes* Baseline SAP-SD SPEC int_rate 2006 VMmark* SPEC_power Specjbb 2005 Up to 63% performance boost over Xeon 5500 9 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, visit http://www.intel.com/performance/resources/limits.htm Source: Intel Internal measurements Jan 2010. See backup for additional details Downbin Performance int_rate fp_rate

Relative Performance and System Power Intel Xeon Processor L5640 Maximize Perf/Watt for Optimized Deployments Dense Servers and Blades High Density Datacenters & Containers Xeon 5570 vs. Xeon L5640 Power and Performance Comparison X5570 4C, 2.93GHz 95W L5640 6C, 2.26GHz 60W Same Performance Up to 30% Lower Power 1 Peak power under load (W) Performance Same performance as a Xeon X5570 SKU, but up to 30% lower system power 1 Source: Fujitsu Performance measurements comparing Xeon L5650 vs X5570 SKUs using SPECint_rate_base2006. See http://docs.ts.fujitsu.com/dl.aspx?id=0140b19d-56e3-4b24-a01e-26b8a80cfe53 and http://docs.ts.fujitsu.com/dl.aspx?id=4af74e10-24b1-4cf8-bb3b-9c4f5f177389 10 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, visit http://www.intel.com/performance/resources/limits.htm

Server Refresh Opportunities in 2010 If you delayed refresh in 2009, then you re not alone IDC believes replacement Approximately cycles being 1 million delayed servers from have had 3-4 their years replacement to every delayed 4-5 by a year. years. Source IDC, 2009 Source: Gartner press release http://www.gartner.com/it/page.jsp?id=1209913 Approximately 1 million 2010 will mark servers an important have had return their to installed replacement base refreshes delayed driven by by an a year. uptick in enterprise budgets, new technological innovations, and a return to economic growth. Source: Gartner press release Source: IDC, February 2010 http://www.gartner.com/it/page.jsp?id=1209913 ~76% of the server install base is still up for refresh Estimated 34% single-core and 42% dual-core 1 single core dual core four core + 2005 2006 2007 2008 2009 Source: Intel. Intel Xeon shipments, 2005-2009. Maintaining Aging Servers Limits Innovation & Growth Now is the Time to Refresh! 1 Source: IDC 2009 Q4 Server Tracker. Install base calculated by using this IDC data and a standard life-cycle distribution, and assumes 4- to 5-year replacement cycle in 2008 and 2009 when the recession hit. 11

Server Refresh: 2010 Single Core Xeon 5600 2005 2010 Efficiency Refresh 15:1 95% Annual Energy Cost Reduction (estimated) 1 rack of Intel Xeon 5600 Based Servers As Low as 5 Month Payback OR 15 racks of Intel Xeon Single Core Servers Performance Refresh 1:1 15 racks of Intel Xeon 5600 Based Servers Up to 15x Performance 8% Annual Energy Costs Estimated Reduction (estimated) Source: Intel measurements as of Feb 2010. Performance comparison using server side java bops (business operations per second). Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. For detailed calculations, configurations and assumptions refer to the legal information slide in backup. 12

Server Refresh: 2010 Dual Core Xeon 5600 2006 2010 Efficiency Refresh 5:1 85% Annual Energy Cost Reduction (estimated) 1 rack of Intel Xeon 5600 Based Servers As Low as 15 Month Payback OR 5 racks of Intel Xeon Dual Core Servers Performance Refresh 1:1 5 racks of Intel Xeon 5600 Based Servers Up to 5x Performance 10% Annual Energy Costs Estimated Reduction (estimated) Source: Intel measurements as of Feb 2010. Performance comparison using server side java bops (business operations per second). Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. For detailed calculations, configurations and assumptions refer to the legal information slide in backup. 13

Benefits of the Intel Xeon Processor 5600 Series Performance Energy Efficiency Virtualization Up to 60% server performance boost over Xeon 5500 1 Up to 15x performance vs. Intel single-core servers 2 Similar performance as Xeon X5570 up to 30% lower power 4 Lower power CPU SKUs (40W) and low power DDR3 memory Robust HW-assisted virtualization Advanced security features for AES-NI and Intel TXT estimated 5-month Refresh ROI Lower Operating Costs Reduce Risks 14 1 Source: Internal Intel measurements for Xeon X5680 vs. Xeon X5570 on BlackScholes* 2 Source: Intel estimates as of Jan 2010. Performance comparison using SPECjbb2005 bops (business operations per second). Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. 3 Source: Internal Intel measurements. Xeon X5670 vs X5570 Measured results on server side java benchmark in conjunction with power consumption across a load line (SPECpower). 4 Source: Fujitsu Performance measurements comparing Xeon L5650 vs X5570 SKUs using SPECint_rate_base2006. See http://docs.ts.fujitsu.com/dl.aspx?id=0140b19d-56e3-4b24-a01e-26b8a80cfe53 and http://docs.ts.fujitsu.com/dl.aspx?id=4af74e10-24b1-4cf8-bb3b-9c4f5f177389 See backup for system configurations on all configurations.

Xeon 7500 Nehalem Generation Intel Microarchitecture Technology Advantages Intel 7500 Chipset Nehalem architecture 8-cores Xeon 7500 Xeon 7500 24MB Shared L3 Cache 64 DIMM slots support up to 1 terabyte of memory (4 sockets) Xeon 7500 Xeon 7500 72 PCIe Gen2 lanes Scaling from 2-256 sockets Intel Virtualization Technologies PCI Express* 2.0 Memory Intel Scalable Memory Buffer Mission Critical Class Reliability features ICH 10/10R Intel 82599 10GbE Controller 15

Xeon 7500 Platform Memory Intel Scalable Memory Buffers enable higher memory capacity 8 DDR3 channels per socket 16 DDR3 DIMMs per socket Supports16GB DDR3 DIMMs (1TB with 4 sockets) Intel Scalable Memory Interconnects enable matching higher bandwidth* Xeon 7500 7500/6500 Memory Per Socket 4 SMI channels 8 DDR3 channels (4 x 2) SMB Memory types supported: 1066MHz DDR3 Registered (RDIMM); Single-rank (SR), dual-rank (DR), quad-rank (QR) Up to 16 DIMMS slots Actual operational memory speed depends on specific processor capabilities (see Xeon 7500 SKU stack for max SMI link speeds by part): 6.4GT/s SMI link speed capable parts support up to 1066Mhz 5.86GT/s SMI link speed capable parts support up to 978Mhz^ 4.8GT/s SMI link speed capable parts support up to 800Mhz *Up to 8X vs. Xeon 7400 per Intel internal benchmark measurement ^Example: 1066MHz memory actually operates at 978Mhz 16

Frequency Core0 Core1 Core8 Core0 Core1 Core8 Core0 Core1 Performance Enhancements Intel Xeon 7500 Series Processor Intel Turbo Boost Technology Increases performance by increasing processor frequency and enabling faster speeds when conditions allow Intel Hyper-threading Technology Increases performance for threaded applications delivering greater throughput and responsiveness <8C Turbo Normal 8C Turbo All cores operate at rated frequency All cores operate at higher frequency Fewer cores may operate at even higher frequencies Higher Performance on Demand Higher Performance For Threaded Workloads 17

Intel Xeon Processor 7500/6500 Series Product Options 8 and 4 socket/ Scalable Usage Advanced Standard Processor Number/Freq X7560 (2.26GHz) Max Scale Glueless 8 skt Cores/ Threads 8/16 QPI Max Mem ** Cache Speed Speed TDP Turbo HT 24M 6.4 GT/s 1066 MHz 130W +3 X7550 (2 GHz) 8 skt 18M 6.4 GT/s 1066 MHz 130W +3 X7542 (2.66GHz) 8 skt 6/6 18M 5.86 GT/s 1066(978) MHz + 130W +1 No E7540 (2 GHz) 8 skt 6/12 18M 6.4 GT/s 1066 MHz 105W +2 E7530 (1.86GHz) 4 skt 12M 5.86 GT/s 1066(978) MHz + 105W +2 Basic E7520 (1.86GHz) 4 skt 4/8 18M 4.8 GT/s 800 MHz 95W No Turbo Low Voltage 2 socket/ Scalable L7555 (1.86 GHz) 8 skt 8/16 24M 5.86 GT/s 1066(978) MHz + 95W +5 L7545 (1.86 GHz) 8 skt 6/12 18M 5.86 GT/s 1066(978) MHz + 95W +5 Usage Processor Number/Freq Max Scale Glueless Cores/ Threads Cache QPI Speed Max Mem Speed TDP ** Turbo HT Advanced X6550 (2 GHz) 2 skt 8/16 18M 6.4 GT/s 1066 MHz 130W +3 Standard E6540 (2 GHz) 2 skt 6/12 18M 5.86 GT/s 1066(978) MHz + 105W +2 Basic E6510 (1.73 GHz) 2 skt only 2 skt*** 4/8 12M 4.8 GT/s 800 MHz 105W No Turbo Advanced features available on higher end processors + 1066 Mhz frequency runs at an effective frequency of 978 Mhz when run at 5.86GHz SMI link speed Max Scale Glueless: Scaling capability refers to maximum supported number of CPUs in a glueless Boxboro-EX platform (e.g. 8 skt means this SKU can be used to populate up to 8 sockets in a single system) **Max Turbo Boost frequency based on number of 133 MHz increments above base freq (+2 = 0.266 GHz, +3 = 0.400 GHz) ***E6510 may not be scaled above 2 sockets even with a customer node controller 18

Product Transition- Xeon 7400 to Xeon 7500 4S Rack Low Power 8S/4S Rack Low Power 2S X7460 2.66 /16M 130W L7455 2.13/12M 65W Advanced 8 cores/scalable Largest Cache Advanced Reliability Turbo X7560 2.26/24M X7550 2/18M 130W 130W L7555 1.86/24M 95W X6550 2 /18M 130W E7450 2.40/12M 90W L7545 1.86/18M 95W E7440 2.40/16M E7430 2.13/12M 90W 80W L7445 2.53/12M L7345 2.53/8M 50W 50W Standard 6 cores/scalable Higher Cache Advanced Reliability Turbo E7540 2/18M E7530 1.86/12M 105W 105W X7542 1 2.66/18M 130W Freq. Optimized E6540 2/18M 105W E7420 2.13 /8M E7310 1.60/4M 19 90W 80W E7220 2.93/8M 80W Xeon 7400 (6 cores) Xeon 7400 (4 cores) Xeon 7300 (Tigerton, 2 & 4 cores) Basic 4 cores/scalable High Cache Advanced Reliability E7520 1.86/18M 95W 1 SKU for HPC & legacy frequency-sensitive applications Xeon 7500/6500 (8 cores) Xeon 7500/6500 (6 cores) Xeon 7500 (4 cores) E6510 1.73/12M 105W 2S only, not scalable

NEW 2-socket Expandable Platform Efficient Performance Class Expandable Class Xeon(R) 5600 Xeon(R) 5600 Xeon 7500 Xeon 7500 IO HUB Xeon 5600 IO HUB Xeon 7500 / 6500* Frequency Sensitive Performance Thread Sensitive Low Power Efficiency High Low Memory Capacity High Good Reliability Best Virtualization driving need for compute, memory & reliability *Use X7560 for highest performance 20

Intel Xeon Processor X7560 Intel Xeon Processor 7500 series-based Servers 4S Standard Benchmarks Performance Summary Memory Bandwidth (GB/s) SPECfp*_rate_base2006 VMmark* SAP* SD 2-tier SPECjbb*2005 TPC Benchmark* E SPECint*_rate_base2006 SPECjAppServer*2004 Xeon X7460 Baseline HPC Consolidation 1.0 2.77 2.64 2.51 3.42 3.18 3.50 3.82 3x average improvement Xeon X7460 = Intel Xeon processor X7460 (16M Cache, 2.66GHz, 1066MHz FSB, formerly codenamed Dunnington) Xeon X7560 = Intel Xeon processor X7560 (24M Cache, 2.26GHz, 6.40GT/s Intel QPI, formerly codenamed Nehalem-EX) 8.24 Relative results Higher is better Average of 3x performance improvement over 7400 series across a range of benchmarks Source: Best published / submitted results comparison of best 4-socket Xeon X7460 and X7560 models as of March 26, 2010. See previous Broad Performance Claim foil and notes for more information. 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, visit http://www.intel.com/performance/resources/limits.htm 21 Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. Copyright * 2010, Other Intel names Corporation. and * Other brands names may and be brands claimed may be as claimed the as property the property of of others. Copyright 2010, Intel Corporation.

4S/64T Intel Xeon Processor X7560 (24M Cache, 2.26 GHz, 6.40 GT/s Intel QPI) Intel Xeon Processor 7500 4S Enterprise Standard Benchmarks Comparison to 2S Xeon 5600 Series SPECfp*_rate_base2006 SPECjbb*2005 SPECjAppServer*2004 SAP* SD 2-tier SPECmpi*2007 SPECint*_rate_base2006 VMmark* SPECompM*2001 Xeon X5680 Baseline (6C, 12M Cache, 3.33GHz) 1.00 2.19 2.17 2.13 2.12 2.11 2.04 2.01 1.95 Relative results Higher is better Xeon 7500 and Xeon 5600 benchmark results represent best top-bin published results as of 29 March 2010. See notes for details. Up to 2.2x the performance of Xeon 5600 22 Source: Best published results on SPEC.org or VMware.com as of 29 March 2010. 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, visit http://www.intel.com/performance/resources/limits.htm Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. Copyright 2010, Intel Corporation. * Other names and brands may be claimed as the property of others.

Advanced RAS Delivers Value For IT Protects Your Data Reduces circuit-level errors Detects data errors across the system Limits the impact of errors Parity Checking And ECC Memory Thermal Throttling Memory Demand & Patrol Scrubbing Corrupt Data Containment Mode Viral Mode Intel QPI Protocol Protection Via CRC (8bit Or 16bit Rolling) Increases Availability Heals failing data connections Supports redundancy and failover for key system components Recovers from uncorrected data errors MCA Recovery With OS Support Intel SMI Lane Failover Intel SMI Clock Fail Over Intel SMI & QPI Packet Retry QPI Clock Fail Over QPI Self-healing SDDC Plus Random Bit Error Recovery Memory Mirroring Memory DIMM And Rank Sparing Dynamic CPU And Memory Migration Minimizes Planned Downtime Helps predict failures before they happen Maintain partitions instead of systems Proactively replace failing components Electronically Isolated (Static) Partitioning MCA Error Logging (CMCI) With OS Predictive Failure Analysis Memory Board Hot Add/Remove OS Memory On-lining* CPU Board Hot Add At QPI OS CPU On-lining Support for Highly Available System Deployments 23

RAS in a Virtualised World Blue screen System keeps running App App OS MCA App App OS VMM App App OS VMM logs error and bug checks CPU signals MCA to VMM with processor context corrupted bit (PCC) logged App App OS MCA App App OS VMM App App OS VMM logs error, terminates affected VM CPU signals MCA to VMM, with PCC bit cleared and address of location with bad data CPU Patrol Scrubber 2xECC error Memory DRAM Patrol Scrubber finds 2xECC error (but SW has not read this data yet) CPU Patrol Scrubber 2xECC error Memory DRAM Patrol Scrubber finds 2xECC error (but SW has not read this data yet) System without error recovery support System with error recovery support 24

Server Refresh Benefits Single Core Xeon 7500 2005 2010 Performance Refresh 1:1 20 Racks of Intel Xeon 7500 Based Servers Up to 20x Performance (Estimated) OR 20 Racks of Intel Xeon Single Core Servers Efficiency Refresh 20:1 1 Rack of Intel Xeon 7500 Based Servers Up to 92% Annual Energy Costs Reduction (Estimated) As Little as 12 months Payback (Estimated) Source: Intel measurements as of March 2010 of Xeon 7500 and single-core 4-socket systems. Performance comparison using SPECint_rate_base2006. Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. For detailed calculations, configurations and assumptions refer to the legal information slide in backup. 25

Intel Xeon Processor 7500 Series Benefits Scalable Performance Up to 3.7X performance boost over Xeon 7400 1 2 terabyte of memory (8 sockets) 2 2-256 sockets 8X memory bandwidth 3 Over 20x performance vs. older, single-core servers 4 Flexible Virtualization I/O Virtualization Intel VT Flex- Migration Assist for virtualization pool investment protection Advanced Reliability Over 20 new RAS features 7 Machine Check Architecturerecovery Recover from fatal errors Biggest Performance Leap Ever for Xeon 5 12 Month Refresh ROI 4 Higher VM Density and Investment Protection Mission Critical Reliability 26 1. up to 3.7X over Xeon 7400 vconsolidate using a leading VM vendor. See performance foil for details 2. 8 socket system with 128 DIMM slots populated with 16GB DDR3 DIMMs 3. 8X per Intel internal memory BW measurement 3.2.10 4. Estimate of Xeon 7500 vs older single core 4socket servers. See 20:1 Refresh Foil for details 5. Per published history of Intel Xeon product performance 6.See RAS list for new features

Linked Foils 27

Modular Platform Drives Innovation Wide Range of Xeon 7500 Platforms Brought to Market 2-socket 2+2+2+2 (8S) 4S (64DIMMs) 4+4 (8S) Add l configs via OEM-specific scaling tech 2+2+ (up to 256s) 2+2 (4S) 4S (32DIMMs) Xeon 7500 CPU Socket Intel QuickPath Interconnect I/O Hub Memory 3rd partry Node Controller (non-intel) OEM interconnect... Huge variety of systems available for optimized choice 28 * Other * Other names names and brands brands may be may claimed be claimed as the as property the property of others. of others. Copyright Copyright 2010, 2010, Intel Corporation. Intel Corporation.

NEW 2-socket Expandable Platform Efficient Performance Class Expandable Class Xeon(R) 5600 Xeon(R) 5600 Xeon 7500 Xeon 7500 IO HUB Xeon 5600 IO HUB Xeon 7500 / 6500* Frequency Sensitive Performance Thread Sensitive Low Power Efficiency High Low Memory Capacity High Good Reliability Best Virtualization driving need for compute, memory & reliability *Use X7560 for highest performance 29

Intel Xeon Processor 5600 Series Building on Xeon 5500 Leadership Capabilities Intel Xeon 5600 Intel Xeon 5600 Intel SSDs Intel 10Gbe 1 Lower Power Dramatic performance gains Intel X25-E SSDs ICH 9/10 Intel 82599 10GbE Controller >5x port count reduction Up to 4.5x power per Gb reduction End-to-end platform power optimization 1 Based on Intel internal results. Actual results may vary significantly based on workload and product configurations. See backup for more details on the results. Back 30

Beyond the Processor Intel End-to-End Power Optimization Processor Optimization 32nm Process Technology 40W CPU SKUs and low power DDR3 support Intelligent Power Technology Intel Xeon 5600 (Xeon 5600-EP) Impact Higher performance/watt Lower system power System level power control via Intel Node Manager Network Optimizations Unified networking Eliminate switches/cables Scalable with multi-core CPUs Intel 10GbE Impact 1 Up to 10X increase IO bandwidth >5X port count reduction Up to 4.5X power per Gb reduction Solid State Drives Lower power consumption No moving parts Dramatic performance increases Intel X25-M SSDs Impact 1 Up to 46X lower power Lower TCO (fewer drives) Up to 6X read perf improvement 1 Based on Intel internal results. Actual results may vary significantly based on workload and product configurations. See backup for more details on the results. Back 31

Security Challenges Continue to Mount in the Data Center Where are my apps and services? New usage models are blurring datacenter perimeters Where are attacks coming from? Shift in types of attack targeting the platform, not just software What are the regulatory requirements? HIPPA, SOX, PCI, etc. requiring security enforcement & auditing Security is a top priority for CIOs in 2010 1 1 Source Ziff Davis Enterprise Top 10 CIO Priorities for 2010. http://www.baselinemag.com/c/a/it-management/cio-priorities-for-2010-706071/ Back 32

Hardware Based Isolation and Security IT Concerns with Software-Only Virtualization Protected Virtual Machines with Intel TXT and Intel VT Virtual Machine A Security Software OS Risk 2 Virtual Machine B Apps OS Risk 3 Virtual Machine B Apps OS 2 Virtual Machine B Apps OS Virtual Machine Monitor 3 Virtual Machine Monitor Risk 1 Typical Business PC 1 Intel Trusted Execution Technology Intel Virtualization Technology Trusted Platform Module v1.2 Risk 1 OS-based security software has no visibility into VMM integrity 1 Intel TXT enables launch of VMM into a known, expected state Risk 2 Isolation of sensitive data in virtual machines is enforced by software alone 2 Isolation of sensitive data in virtual machines is enforced by software alone Risk 3 Disorderly shutdown of a virtual machine can expose data 3 Disorderly shutdown of a virtual machine can expose data Back 33

Intel TXT Platform Requirements Intel TXT requires ALL of the following components: Intel Xeon 5600 Series Processor Intel 5500 or 5520 chipset (C2 stepping) Trusted Platform Module (TPM) Enabled platform BIOS Enabled Hypervisor or Operating System A platform without any of these components will boot normally, but without Intel TXT trusted launch support Back 34

Intel TXT Ecosystem and Enabling OEM Platforms: Intel expects a number of leading OEMs and Intel s EPSD to deliver TXT-capable platforms in 2010 Unlikely that OEM platforms will have full TXT capability at Intel Xeon 5600 processor launch Intel will support several OEMs announcing platforms that will be TXT Ready with all required platform HW components at launch, with a TXT-enabled BIOS upgrade post-launch Software: Ecosystem enablement spread through 2010-2011 Pending completion of enablement activity on qual platforms and OEM system validation Two Hypervisor vendors targeting Q3/Q4 10 releases supporting TXT, with 2+ additional Hypervisor / OS vendors pending 1H 11 Back 35

BACKUP 36

Processor Transition Matrix Which Xeon 5600 Product is Right for Me? Price Brand 1 Series, SKU 5100 (Dual-core Xeon) Current Products 5300 (65nm Quadcore Xeon) 5400 (45nm, Quadcore Xeon) 5500 (45nm, Quadcore Nehalem) NEW Product Xeon 5600 Xeon 5600-EP denotes Six-core SKU >$800 5160 (3.00GHz) X5365 (3.00GHz) X5470 (3.33GHz) X5460 (3.16GHz) E5450 (3.00GHz) X5450 (3.00GHz) X5570 (2.93GHz) X5560 (2.80GHz) X5550 (2.66GHz) X5680 (3.33GHz) X5670 (2.93GHz) X5660 (2.80GHz) X5650 (2.66GHz) Advanced Performance oriented customers seeking highest functionality and optimal server ROI $301- $799 5150 (2.66GHz) 5140 (2.33GHz) 5130 (2.00GHz) X5355 (2.66GHz) E5345 (2.33GHz) E5335 (2.00GHz) E5440 (2.83GHz) E5430 (2.66GHz) E5420 (2.50GHz) E5540 (2.53GHz) E5530 (2.40GHz) E5520 (2.26GHz) E5640 (2.66GHz) E5630 (2.53GHz) E5620 (2.40GHz) Standard Customers seeking a mix of performance, value, and advanced features <$300 5120 (1.86GHz) 5110 (1.60GHz) E5320 (1.86GHz) E5310 (1.60GHz) E5410 (2.33GHz) E5405 (2.00GHz) E5506 (2.13GHz) E5504 (2.00GHz) E5502 (1.86GHz) E5507 (2.26GHz) E5506 (2.13GHz) E5503 (2.00GHz) Basic Cost-sensitive customers seeking basic features Low Power 5148 (2.33GHz) L5335 (2.00GHz) L5320 (1.86GHz) L5310 (1.60GHz) L5530 (2.66GHz) L5420 (2.50GHz) L5410 (2.33GHz) L5530 (2.26GHz) L5520 (2.26GHz) L5506 (2.13GHz) 1 Source: Intel. Pricing reflects 1k unit price to OEMs at time of introduction. Subject to change without notice. L5540 (2.26GHz) L5530 (2.13GHz) L5609 (1.86GHz) Low Power Customer seeking lowest power or use in constrained form factors (blades) Back 37

Intel Xeon Processor 5600 Series Product Options Usage Advanced Standard Processor Number/Freq Cores/ Threads Cache QPI Speed Max Mem Speed TDP Turbo 1 HT X5680 (3.33 GHz) 6/12 12M 6.4 GT/s 1333 MHz 130W +2 X5677 (3.46 GHz) 4/8 12M 6.4 GT/s 1333 MHz 130W +2 X5670 (2.93 GHz) 6/12 12M 6.4 GT/s 1333 MHz 95W +3 X5667 (3.06 GHz) 4/8 12M 6.4 GT/s 1333 MHz 95W +3 X5660 (2.80 GHz) 6/12 12M 6.4 GT/s 1333 MHz 95W +3 X5650 (2.66 GHz) 6/12 12M 6.4 GT/s 1333 MHz 95W +3 E5640 (2.66 GHz) 4/8 12M 5.86 GT/s 1066 MHz 80W +2 E5630 (2.53 GHz) 4/8 12M 5.86 GT/s 1066 MHz 80W +2 E5620 (2.40 GHz) 4/8 12M 5.86 GT/s 1066 MHz 80W +2 L5640 (2.26 GHz) 6/12 12M 5.86 GT/s 1333 MHz 60W +4 Low Voltage L5630 (2.13 GHz) 4/8 12M 5.86 GT/s 1066 MHz 40W +2 L5609 (1.86 GHz) 4/4 12M 4.8 GT/s 1066 MHz 40W N/A Basic [Xeon 5500 based SKUs] E5507 (2.26 GHz) 4/4 4M 4.8 GT/s 800 MHz 80W N/A E5506 (2.13 GHz) 4/4 4M 4.8 GT/s 800 MHz 80W N/A E5503 (2.00 GHz) 2/2 4M 4.8 GT/s 800 MHz 80W N/A 1 Max Turbo Boost frequency based on number of 133 MHz increments above base freq (+2 = 0.266 GHz, +3 = 0.400 GHz) 38

Intel Xeon 5600 Public Pricing 1ku quantities or less Processor Number Frequency Cache Power Cores / Threads 1kU Price Intel Xeon Processor X5680 3.33 GHz 12 MB 130W 6 / 12 $1663 Intel Xeon Processor X5677 3.46 GHz 12 MB 130W 4 / 8 $1663 Intel Xeon Processor X5670 2.93 GHz 12 MB 95W 6 / 12 $1440 Intel Xeon Processor X5667 3.06 GHz 12 MB 95W 4 / 8 $1440 Intel Xeon Processor X5660 2.80 GHz 12 MB 95W 6 / 12 $1219 Intel Xeon Processor X5650 2.66 GHz 12 MB 95W 6 / 12 $996 Intel Xeon Processor E5640 2.66 GHz 12 MB 80W 4 / 8 $774 Intel Xeon Processor E5630 2.53 GHz 12 MB 80W 4 / 8 $551 Intel Xeon Processor E5620 2.40 GHz 12 MB 80W 4 / 8 $387 Intel Xeon Processor E5507 2.26 GHz 4 MB 80W 4 / 4 $276 Intel Xeon Processor E5506 2.13 GHz 4 MB 80W 4 / 4 $219 Intel Xeon Processor E5503 2.00 GHz 4 MB 80W 2 / 2 $188 Intel Xeon Processor L5640 2.26 GHz 12 MB 60W 6 / 12 $996 Intel Xeon Processor L5630 2.13 GHz 12 MB 40W 4 / 8 $551 Intel Xeon Processor L5609 1.86 GHz 12 MB 40W 4 / 4 $440 Intel Xeon Processor E5645 2.40 GHz 12 MB 80W 6 / 12 $958 Intel Xeon Processor L5638 2.00 GHz 12 MB 60W 6 / 12 $958 Intel Xeon Processor L5618 1.86 GHz 12 MB 40W 4 / 8 $530 39

Intel Xeon Processor 5600 Series (Xeon 5600-EP) Power Data Processor Number Rated Freq GHz (TDP) Cores C1E (W) Package C3 (W) Package C6 (W) X5680 3.33 (130) 6 37 32 12 X5677 3.46 (130) 4 37 32 12 X5670 2.93 (95) 6 32 28 10 X5667 3.06 (95) 4 32 28 10 X5660 2.80 (95) 6 32 28 10 X5650 2.66 (95) 6 32 28 10 E5640 2.66 (80) 4 37 33 12 E5630 2.53 (80) 4 37 33 12 E5620 2.40 (80) 4 37 33 12 E5507 2.26 (80) 4 40 33 14 E5506 2.13 (80) 4 40 33 14 E5503 (2C) 2.00 (80) 2 40 33 14 L5640 2.26 (60) 6 25 20 8 L5630 2.13 (40) 4 22 18 8 L5609 1.86 (40) 4 22 18 8 40

Intel Xeon Processor 5500 Series (Nehalem-EP) Power Data Processor Number Rated Freq GHz (TDP) Cores C1E (W) Package C3 (W) Package C6 (W) W5590 3.33 (130) 4 35 30 12 W5580 3.20 (130) 4 35 30 12 W3520 2.66 (130) 4 35 30 12 X5570 2.93 (95) 4 30 26 10 X5560 2.80 (95) 4 30 26 10 X5550 2.66 (95) 4 30 26 10 E5540 2.53 (80) 4 30 26 10 E5530 2.40 (80) 4 30 26 10 E5520 2.26 (80) 4 30 26 10 E5506 2.13 (80) 4 40 35 15 E5504 2.00 (80) 4 40 35 15 E5502 (2C) 1.86 (80) 2 40 35 15 L5530 2.40 (60) 4 22 18 8 L5520 2.26 (60) 4 22 18 8 L5506 2.13 (60) 4 22 18 8 41

Intel TXT: Be Prepared for Emerging Security Threats Processor Platform Software Intel Xeon 5600 Intel Xeon 5600 Intel 5520 Chipset Trusted Platform Module 1.2 VT TPM Intel 5600 Processor (Xeon 5600-EP) BIOS code Firmware OS and VMM support HW-assisted protection against rootkits and other malware threats at launch Complements runtime protections, such as anti-virus and intrusion detection Start future-proofing today for more secure datacenters with Intel TXT-capable platforms 42

Relative Perf / Watt and CPU TDP Power Greater Performance per Watt Greater Datacenter Efficiency Xeon 5500 and 5600 Power and Performance Comparison Up to 40% More Perf /Watt 1 X5570 (4C) 2.93GHz 95W CPU TDP (Watts) X5670 (6C) 2.93GHz 95W Performance/Watt Same CPU TDP Power Up to 40% fewer servers Lower utility and SW costs 43 1 Source: Internal Intel estimates comparing Xeon X5670 vs. X5570 SKUs using SPECpower benchmark. See backup for system configurations. 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, visit http://www.intel.com/performance/resources/limits.htm

Number of Users Decryption Time(us) provisioning time (seconds) Intel Xeon Processor 5600 Series Intel AES-NI Encryption Performance Web Banking Workload (MS IIS/PHP) 1 4.5x 23% Higher is better Database Encryption/Decryption (Oracle 11g) 2-89% Lower is better Full Disk Encryption (McAfee Endpoint Encryption) 3-42% Lower is better 13000 16000 158 17 3500 20 12 WDC w/o encryption NHM w/o encryption WSM NHM w/o Intel IPP WSM NHM WSM 1 System configuration: Windows 2008 R2 x64 Ent. Server. PHP banking sessions /users measured with Intel Xeon X5680 (3.33 GHz) vs Intel Xeon 5160 (3.00 GHz) and Intel Xeon X5570 (2.93 GHz), 24 SSD RAID 0 arrays, TLS_RSA_with_AES_128_CBC_SHA cipher suite. 2 System configuration: Oracle 11g with TDE, time takes to decrypt a 5.1 million row encrypted table with AES-256 CBC mode on WSM 3.33 GHz optimized with Intel Performance Primitives crypto library (IPP) vs NHM 2.8 GHz without IPP. Timing measured is per 4K of data. 3 System configuration: McAfee Endpoint Encryption for PCs (EEPC) 6.0 package with McAfee epolicy Orchestrator (epo) 4.5 encrypting a 32GB X25E SSD with WSM 3.33 GHz vs. NHM 2.93 GHz. 24GB of memory. 44

Tick-Tock Development Model Sustained Microprocessor Leadership Tick Tock Tick Tock Tick Tock Tick Tock 65nm 45nm 32nm 22nm Intel Core Microarchitecture Nehalem Microarchitecture Sandy Bridge Microarchitecture Intel Xeon 5600 Intel s first 32nm SERVER processor with 6 cores and 12 threads 45

Intel Xeon Processor 5600 Series 32nm Westmere High-k + Metal Gate Transistors Memory Controller M i s c I O & Q P I Core Core Core Core Core Core Shared L3 Cache ** Q u e u e & U n c o r e Shared L3 Cache ** M i s c I O & Q P I Transistor count: 1.17B Die size: 248mm 2 46

Performance Backup 47

Performance Claim Backup 60% Performance, 40% Perf/Watt, and 30% Lower Power Up to 1.6x performance compared to Xeon 5500 series claim supported by a CPU intensive benchmark (Blackscholes). Intel internal measurement. (Feb 25, 2010) Configuration details: - Blackscholes* Baseline Configuration and Score on Benchmark:- Intel pre-production system with two Intel Xeon processor X5570 (2.93 GHz, 8 MB last level cache, 6.4 GT/sec QPI), 24GB memory (6x4GB DDR3-1333), 4 x 150GB 10K RPM SATA RAID0 for scratch, Red Hat* EL 5 Update 4 64-bit OS. Source: Intel internal testing as of February 2010. SunGard v3.0 source code compiled with Intel v11.0 compiler. Elapsed time to run benchmark: 18.74 seconds. New Configuration and Score on Benchmark:- Intel pre-production system with two Intel Xeon processor X5680 (3.33 GHz, 12 MB last level cache, 6.4 GT/sec QPI), 24GB memory (6x4GB DDR3-1333), 4 x 150GB 10K RPM SATA RAID0 for scratch, Red Hat* EL 5 Update 4 64-bit OS. Source: Intel internal testing as of February 2010. SunGard v3.0 source code compiled with Intel v11.0 compiler. Elapsed time to run benchmark: 11.51 seconds. Up to 40% higher performance/watt compared to Intel Xeon Processor 5500 Series claim supported by performance results on a server side java benchmark in conjunction with power consumption across a load line. Intel internal measurement (Jan 15, 2010) Baseline platform: Intel preproduction server platform with two Quad-Core Intel Xeon processor X5570, 2.93 GHz, 8MB L3 cache, 6.4QPI, 8GB memory (4x2GB DDR3-1333), 1 PSU, Microsoft Windows Server 2008 Enterprise SP2. Intel internal measurement as of January 15,2010. New platform: Intel preproduction server platform with two six-core Intel Xeon processor X5670, 2.93 GHz, 12MB L3 cache, 6.4QPI, 8GB memory (4x2GB DDR3-1333), 1 PSU, Microsoft Windows Server 2008 Enterprise SP2. Intel internal measurement as of January 15, 2010. Intel Xeon processor 5600 series with Intel microarchitecture Nehalem delivers similar performance as previous-generation servers but uses up to 30 percent less power Baseline Configuration and Score on Benchmark: - Fujitsu PRIMERGY RX300 S5 system with two Intel Xeon processor sx5570 (2.93 GHz, 8MB L3, 6.4 GT/s, Quad-core, 95W TDP), BIOS rev. R1.09, Turbo Enabled, HT Enabled, NUMA Enabled, 5 x Fans, 24 GB (6x4GB DDR3-1333 DR registered ECC), 1 x Fujitsu MBD2147RC 147GB 10K RPM 2.5 SAS HDD, 1x800W PSU, SLES 11 (X86_64) Kernel 2.6.27.19-5-default. Source: Fujitsu Performance Lab testing as of Mar 2010. SPECint_rate_base2006 score: 250. http://docs.ts.fujitsu.com/dl.aspx?id=0140b19d-56e3-4b24-a01e-26b8a80cfe53 New Configuration and Score on Benchmark:- Fujitsu PRIMERGY RX300 S6 system with two Intel Xeon processors L5640 (2.26 GHz, 12MB L3, 5.86 GT/s, Hex-core, 60W TDP), BIOS rev R1.00A, Turbo Enabled, HT Enabled, NUMA Enabled, 5 x Fans, 24 GB (6x4GB DDR3-1333 LV DR registered ECC), 1 x Fujitsu MBD2147RC 147GB 10K RPM 2.5 SAS HDD, 1x800W PSU, SLES 11 (X86_64) Kernel 2.6.27.19-5-default. Source: Fujitsu Performance Lab testing as of Mar 2010. SPECint_rate_base2006 score: 250 http://docs.ts.fujitsu.com/dl.aspx?id=4af74e10-24b1-4cf8-bb3b-9c4f5f177389 48

Performance Summary Backup Mainstream Server Benchmarks 42% gain on Single Node server SPECpower*_ssj2008 at similar power level over previous generation processors supported by the following: Baseline Configuration and Score: Referenced as published at 2053 overall ssj_ops/watt http://www.spec.org/power_ssj2008/results/res2009q4/power_ssj2008-20091023-00205.html New Configuration and Score: IBM x3650 M3 was configured with the Intel Xeon Processor X5670 (2.93GHz, 256KB L2 cache per core, 12MB L3 cache per processor 12 cores/2 chips/6 cores per chip) and 12GB of PC3L-10600R(6 x 2GB) memory and ran IBM Java 6 Runtime Environment and Microsoft Windows Server 2008 R2 Enterprise x64 Edition. Score: 2,927 overall ssj_ops/watt. Submitted and in review at www.spec.org 46% gain on SPECjbb2005 supported by the following: Baseline Configuration and Score: 632,425 bops, SPECjbb2005 bops/jvm = 158106 http://www.spec.org/osg/jbb2005/results/res2010q1/jbb2005-20100210- 00803.html New Configuration and Score: Fujitsu PRIMERGY RX300 S6 system with two Intel Xeon processors X5680 (3.33 GHz, 12MB L3, 6.4 GT/s, Hex-core, 130W TDP), Turbo Enabled, HT Enabled, NUMA Enabled, Data Reuse Optimization disabled, all prefetchers disabled, 48 GB (12x4GB DDR3-1333 DR registered ECC), 1 x Seagate 73GB 10K RPM 2.5 SAS HDD, Microsoft Windows Server 2008 R2 Enterprise, IBM J9 VM (build 2.4, JRE 1.6.0 IBM J9 2.4 Windows Server 2008 amd64-64 jvmwa6460sr6-20090923_42924). Source: Fujitsu Performance Lab testing as of Mar 2010. SPECjbb2005 score: bops= 928393, bops/jvm= 154732 http://docs.ts.fujitsu.com/dl.aspx?id=71488796-7a53-46b8-9163-61373214c2ef 27% boost on SAP ERP 6.0 Unicode over previous generation supported by the following: Baseline Configuration and Score: 3800 Number of SAP SD benchmark users http://download.sap.com/download.epd?context=40e2d9d5e00eef7c259ffe6ab54898440c838ded66684afd7d58b23a917f4c0d New Configuration and Score : Fujitsu PRIMERGY RX300 S6 system with two Intel Xeon processor X5680 (3.33 GHz, 12MB L3, 6.4 GT/s, Hex-core, 130W TDP), 88 GB main memory, Windows Server 2008 Enterprise Edition, SQL Server 2008, SAP ERP 6.0 (Unicode). Source: www.sap.com Score: 4860 Number of SAP SD benchmark users 40% gain on SPECint_rate_2006 over previous generation supported by the following: Baseline Configuration and Score: Score: 253 http://www.spec.org/cpu2006/results/res2010q1/cpu2006-20100202-09561.html New Configuration and Score: Dell PowerEdge R710 system with two Intel Xeon processor X5680 (3.33 GHz, 12MB L3, 6.4 GT/s, six-core, 130W TDP), Maximum Performance Power Management mode, Data Reuse Disabled, C1E Enabled, Turbo Enabled, HT Enabled, 48 GB (12x4GB DDR3-1333 registered ECC), 1x146GB 15K RPM SAS HDD, SUSE Linux Enterprise Server 11 (2.6.27.19-5-smp). Source: Submitted to www.spec.org for publication as of March 15 2010. Geomean Score of 12 workloads: 355 42% gain on VMMark* over previous generation supported by the following: Baseline Configuration and Score: Cisco result referenced as published at 25.06 at 17 tiles. For more information see www.vmware.com/files/pdf/vmmark/vmmark-cisco-2010-01-12-b200m1.pdf New Configuration and Score on Benchmark:- Cisco UCS B250 M2 platform with two Intel Xeon processor X5680 (3.33 GHz, 12MB L3, 6.4 GT/s, 6-core, 130W TDP), Turbo Enabled, HT Enabled, 192GB memory (48x4GB DDR3 1333), EMC CLARiiON CX4-240 storage system with 25x73GB SSD, 20 x 450GB 15K RPM, 5 x 300GB 15K RPM, VMware vsphere 4,0 U1 Source: www.cisco.com. Score of 35.83@26 tiles. For more information see: www.cisco.com/en/us/prod/ps10265/at_work_promo.html#~industry_benchmarks. 49

Performance Summary Backup Technical Computing Benchmarks STREAM: Baseline Configuration and Score on Stream-MP Benchmark:- Intel pre-production system with two Intel Xeon processor X5570 (2.93 GHz, 12MB L3, 6.4 GT/s, Quad-core, 130W TDP), C3 Disabled, C6 Enabled, Turbo Disabled, HT Disabled, NUMA Enabled, 24 GB (6x4GB DDR3-1333 DR registered ECC), 1x150GB 10K RPM SATA HDD, Red Hat EL5-U4 kernel 2.6.18-164.el5 experimental.8). Source: Intel internal testing TR1012. Score of workloads: 36588.0 MB/s New Configuration and Score on Stream-MP Benchmark:- Intel pre-production system with two Intel Xeon processor X5680 (3.33 GHz, 12MB L3, 6.4 GT/s, Hex-core, 130W TDP), C3 Disabled, C6 Enabled, Turbo Disabled, HT Disabled, NUMA Enabled, 24 GB (6x4GB DDR3-1333 DR registered ECC), 1x150GB 10K RPM SATA HDD, Red Hat EL5-U4 kernel 2.6.18-164.el5 experimental.8). Source: Intel internal testing as of Feb 2010. Score of workloads: MB/s. CAE: Baseline Configuration and Score on CAE Vertical: 2-socket server with Intel Xeon processor X5570 (2.93 GHz, 8MB LLC, 6.4 GT/S QPI, 95W TDP), Turbo Enabled, HT Enabled or Disabled for best performance on each application, 24 GB memory (6x4GB DDR3-1333 registered ECC), 4x150GB 10K RPM SATA HDD RAID0 for scratch, Red Hat* EL 5.4 64-bit OS (2.6.18-164-el5). Source: Intel internal testing as of January 2010. Geometric mean score for nine applications New Configuration and Score on CAE Vertical: Intel pre-production 2-socket server with Intel Xeon processor X5680 (3.33 GHz, 12MB LLC, 6.4 GT/S QPI, 130W TDP), Turbo Enabled, HT Enabled or Disabled for best performance on each application, 24 GB memory (6x4GB DDR3-1333 registered ECC), 4x150GB 10K RPM SATA HDD RAID0 for scratch, Red Hat 5.4 64-bit OS (2.6.18-164-el5). Source: Intel internal testing as of January 2010. Geometric mean score for nine applications SPECfp_rate: 25% gain on SPECfp_rate_base2006 over previous generation supported by the following: Baseline Configuration and Score: Score 197, Referenced as published at http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090511-07354.html New Configuration and Score: Dell PowerEdge R710 system with two Intel Xeon processor X5680 (3.33 GHz, 12MB L3, 6.4 GT/s, six-core, 130W TDP), Maximum Performance Power Management mode, Data Reuse Disabled, C1E Enabled, Turbo Enabled, HT Enabled, 48 GB (12x4GB DDR3-1333 registered ECC), 1x146GB 15K RPM SAS HDD, SUSE Linux Enterprise Server 11 (2.6.27.19-5-smp). Source: Submitted to www.spec.org for publication as of March 15 2010. Geomean Score of 17 workloads: 248. LIFE SCIENCES: Baseline Configuration and Score on Life Sciences Vertical: 2-socket server with Intel Xeon processor X5570 (2.93 GHz, 8MB LLC, 6.4 GT/S QPI, 95W TDP), Turbo Enabled, HT Enabled or Disabled for best performance on each application, 24 GB memory (6x4GB DDR3-1333 registered ECC), 4x150GB 10K RPM SATA HDD RAID0 for scratch, Red Hat* EL 5.4 64-bit OS (2.6.18-164-el5). Source: Intel internal testing as of January 2010. Geometric mean score for ten applications New Configuration and Score on Life Sciences Vertical: Intel pre-production 2-socket server with Intel Xeon processor X5680 (3.33 GHz, 12MB LLC, 6.4 GT/S QPI, 130W TDP), Turbo Enabled, HT Enabled or Disabled for best performance on each application, 24 GB memory (6x4GB DDR3-1333 registered ECC), 4x150GB 10K RPM SATA HDD RAID0 for scratch, Red Hat 5.4 64-bit OS (2.6.18-164-el5). Source: Intel internal testing as of January 2010. Geometric mean score for ten applications LINPACK: Baseline Configuration and Score on Linpack Benchmark:- Supermicro* pre-production system with two Intel Xeon processor X5570 (2.93 GHz, 8MB L3, 6.4 GT/s, Quad-core, 95W TDP), BIOS rev 02/23/2009, C3 Disabled, C6 Enabled, Turbo Enabled, HT Disabled, NUMA Enabled, 24 GB (6x4GB DDR3-1333 DR registered ECC), 1x150GB 10K RPM SATA HDD, Red Hat EL5-U3 kernel 2.6.18-128.el5 for x86_64). Source: Intel internal testing TR1011A. Score of workloads: 91 GFlops. New Configuration and Score on Linpack Benchmark:- Supermicro* pre-production system with two Intel Xeon processor X5680 (3.33 GHz, 12MB L3, 6.4 GT/s, Hexcore, 130W TDP), BIOS rev 01/06/2010, C3 Disabled, C6 Enabled, Turbo Enabled, HT Disabled, NUMA Enabled, 24 GB (6x4GB DDR3-1333 DR registered ECC), 1x150GB 10K RPM SATA HDD, Red Hat EL5-U4 kernel 2.6.18-164.el5 experimental.8). Source: Intel internal testing as of Feb 2010. Score of workloads: 146 GFlops. BLACKSCHOLES*: Baseline Configuration and Score on Benchmark:- Intel pre-production system with two Intel Xeon processor X5570 (2.93 GHz, 8 MB last level cache, 6.4 GT/sec QPI), 24GB memory (6x4GB DDR3-1333), 4 x 150GB 10K RPM SATA RAID0 for scratch, Red Hat* EL 5 Update 4 64-bit OS. Source: Intel internal testing as of February 2010. SunGard v3.0 source code compiled with Intel v11.0 compiler. Elapsed time to run benchmark: 18.74 seconds. New Configuration and Score on Benchmark:- Intel pre-production system with two Intel Xeon processor X5680 (3.33 GHz, 12 MB last level cache, 6.4 GT/sec QPI), 24GB memory (6x4GB DDR3-1333), 4 x 150GB 10K RPM SATA RAID0 for scratch, Red Hat* EL 5 Update 4 64-bit OS. Source: Intel internal testing as of February 2010. SunGard v3.0 source code compiled with Intel v11.0 compiler. Elapsed time to run benchmark: 11.51 seconds. 50