AMD Opteron™ 6200 Series Processor Quick Reference Guide The world’s first 16-core x86 processor, delivering a rich mix of performance, scalability and efficiency > for today’s highly threaded computing environments1. AMD Opteron™ 6200 Series Processor KEY FEATURES >>New Core Architecture — drives more core density1 and greater throughput2 >>AMD Turbo CORE Technology — allows processors to independently boost their clock speeds, scaling frequency up 500MHz-1GHz+ automatically to respond > to the need for more application performance3 >>Flex FP — delivers 256-bit floating point processing > with more throughput for both 128-bit and 256-bit technical applications4 >>AMD Virtualization™ (AMD-V™) Technology 2.0 — > heightens virtualization efficiency with new enhancements to the AMD-V™ suite of virtualization to optimize data center rack space and help minimize management tasks END USER BENEFITS Unleash unprecedented performance of highly threaded applications through massive, industry-leading core density >>A new 16-core architecture for the next generation of scalable business critical applications, featuring an innovative design that drives more core density and greater throughput with > up to 160% more cores than the competition5 >>The modular design features up to 33% more cores6 and greater throughput7 in a smaller silicon footprint8 than AMD’s current products, bringing high performance throughput for scalable computing environments like virtualization, HPC, > web/cloud clusters and database applications >>Performance features designed for leadership across threaded workloads and price/performance leadership across even more workloads >>Straight-through computing helps ensure that there are no bottlenecks or compromises — when the workload grows up to 16 tasks each has their own dedicated core, with maximum memory and I/O speed available on every processor SKU regardless of the price >>Straight-through computing ensures the memory and I/O architecture supports the power of more cores. >>Break HPC barriers with the ability to scale to higher core counts with fewer nodes9. Bring unparalleled efficiency to your processing, power and financial budgets >>Optimize data center floor space and power requirements with up to 16 cores, dramatically boosting processing efficiency10 allowing enterprises to do more work with fewer systems.11 >>Straight-through computing capability gives computeintensive applications like database, HPC and Java dedicated processing resources. >>New instructions help streamline software development by minimizing code differences between platforms and help more efficiently process applications, especially media encoding and streaming, that utilizes complex > SSE codes. Today’s highly threaded applications demand more scalability and the AMD Opteron™ 6000 Series platform delivers the most cores in the industry5 >>As user loads increase for web/cloud, database applications or virtualized environments, AMD Opteron™ 6200 Series processors can continue to seamlessly handle growing workloads and allow balanced scaling up to 16 cores per processor. 1 Comparison of 16-core AMD Opteron™ 6200 Series processor with 8-core Intel Xeon E5-2600 Series processor and 10-core Intel Xeon E7 Series processors according to www.intc.com/pricelist.cfm as of 4/12/12. SVR-30 2TPC and TPC-C are trademarks of the Transaction Processing Performance Council. The results stated above reflect results published on http://www.tpc.org as of November 28, 2011. The comparison presented above is based on the best performing two-socket servers using AMD Opteron™ processor Models 6282 SE and 6176 SE, operating at each processor’s default frequency. For the latest TPC-C results, visit http://www.tpc.org. Performance (tpmc) = 1,207,982, 2 x AMD Opteron™ processors Model 6282 SE: http://www.tpc.org/tpcc/results/tpcc_result_detail.asp?id=111111501. Performance (tpmc) = 705,652, 2 x AMD Opteron™ processors Model 6176 SE: http://www.tpc.org/tpcc/results/tpcc_ result_detail.asp?id=110040801. SVR-16 3 AMD Opteron 6200 Series processors experience all core boost of up to 500 MHz (P2 base to P1 boost state) and up to 1.3 GHz max turbo boost (half or fewer cores boost from P2 to P0 boost state). SVR-27 4Based on two 2.7 GHz AMD Opteron 6200 Series processors, each with 16 128-bit FPUs each able to handle 4 instructions per cycle or 8 256-bit Flex FP units handling 8 instructions per cycle. SVR-29 5Comparison of 16-core AMD Opteron™ 6200 Series processor with 8-core Intel Xeon E5-2600 Series processor and 10-core Intel Xeon E7 Series processors according to www.intc.com/pricelist.cfm as of 4/12/12. SVR-30 6Comparison of 12-core AMD Opteron 6100 Series processors and 16-core AMD Opteron 6200 Series processors. SVR-5 7 TPC and TPC-C are trademarks of the Transaction Processing Performance Council. The results stated above reflect results published on http://www.tpc.org as of November 28, 2011. The comparison presented above is based on the best performing two-socket servers using AMD Opteron™ processor Models 6282 SE and 6176 SE, operating at each processor’s default frequency. For the latest TPC-C results, visit http://www.tpc.org. Performance (tpmc) = 1,207,982, 2 x AMD Opteron™ processors Model 6282 SE: http://www.tpc.org/tpcc/results/tpcc_result_detail.asp?id=111111501. Performance (tpmc) = 705,652, 2 x AMD Opteron™ processors Model 6176 SE: http://www.tpc.org/tpcc/results/tpcc_ result_detail.asp?id=110040801. SVR-16 8Based on AMD Opteron 6100 Series processor at 2x 346 mm2 vs. AMD Opteron 6200 Series processor at 2x 316mm2. SVR-36 9Twelve 2P servers with AMD Opteron 6176 SE @ 2.3 GHz offer 2650 GFLOPs total performance. Nine 2P servers with AMD Opteron 6276 @ 2.3 GHz offer 2650 GFLOPs for the same amount of total processing power in 25% fewer servers. SVR-35 10According to AMD internal engineering estimates, the AMD Opteron™ 6200 Series processors will deliver up to 35% greater throughput in the same general power/thermal envelopes as AMD Opteron 6100 Series processors. SVR-34 11 Twelve 2P servers with AMD Opteron 6176 SE @ 2.3 GHz offer 2650 GFLOPs total performance. Nine 2P servers with AMD Opteron 6276 @ 2.3 GHz offer 2650 GFLOPs for the same amount of total processing power in 25% fewer servers. SVR-35 AMD Opteron™ 6200 Series Processor Quick Reference Guide AMD Opteron™ 6200 Series Processor Product Specifications AMD Opteron™ 6200 Series Processor Product Specifications Model Number Core Count AMD Turbo All Core Core CORE Max Frequency Boost Frequency Frequency North Bridge L2 Cache 6284 SE* 16 2.7 GHz 3.1 GHz 3.4 GHz 2.0 GHz 8 x 2 MB 16 MB 140W 6282 SE 16 2.6 GHz 3.0 GHz 3.3 GHz 2.0 GHz 8 x 2 MB 16 MB 140W 6278* 16 2.4 GHz 2.7 GHz 3.3 GHz 2.0 GHz 8 x 2 MB 16 MB 6276 16 2.3 GHz 2.6 GHz 3.2 GHz 2.0 GHz 6274 16 2.2 GHz 2.5 GHz 3.1 GHz 6272 12 2.1 GHz 2.4 GHz 6262 HE 16 1.6 GHz 6238 12 6234 Cache Sizes Total Cache: 32MB (16 core), 28MB (12 core), > 24MB (8 core), 20 MB (4 core) L1 Cache: 16KB/core + 64KB instruction/module L2 Cache: 1MB (per core) L3 Cache: 16MB (per socket) 115W Process Technology 32-nanometer SOI (silicon-on-insulator) technology 8 x 2 MB 16 MB 115W HyperTransport™ Technology 3.0 4X HT3 links with peak bandwidth of 6.4 GT/s per link 2.0 GHz 8 x 2 MB 16 MB 115W Memory 3.0 GHz 2.0 GHz 8 x 2 MB 16 MB 115W Integrated DDR3 memory controller — Up to 102.4 GB/s memory bandwidth per CPU for Socket G34 2.1 GHz 2.9 GHz 1.8 GHz 8 x 2 MB 16 MB 85W Number of Channels/> Types of Memory Quad Channel support for U/RDDR3 up to DDR3-1600 and ULV (1.25V) RDDR3 up to DDR3-1333 and LRDIMM up to DDR3-1333 2.6 GHz 2.9 GHz 3.2 GHz 2.0 GHz 6 x 2 MB 16MB 115W Die Size 316mm2 12 2.4 GHz 2.7 GHz 3.0 GHz 2.0 GHz 6 x 2 MB 16 MB 115W Packaging Socket G34 — 1944-pin organic Land Grid Array (LGA) 6220 8 3.0 GHz 3.3 GHz 3.6 GHz 2.0 GHz 4 x 2 MB 16 MB 115W 6212 8 2.6 GHz 2.9 GHz 3.2 GHz 2.0 GHz 4 x 2 MB 16 MB 115W 6204 4 2.3 GHz N/A N/A 2.0 GHz 2 x 2 MB 16 MB 115W L3 Cache TDP *New SKUs available with Speed Bump Launch 2012 AMD SP5100 Southbridge > Product Specifications AMD SR5650, SR5670, SR5690 I/O Hub Product Specifications Model Number Processor Interface PCI Express® SR5650 HyperTransport™ 3.0 technology (5.2GT/s) SR5670 HyperTransport 3.0 technology (5.2GT/s) SR5690 HyperTransport™ 3.0 technology (5.2GT/s) v2.0 ™ v2.0 v2.0 Number of PCIe® Ports/ Engines 22 lanes/ > 8 engines 30 lanes/ > 9 engines 42 lanes/> 11 engines Max TDP/ Idle (w/c1e) Process Technology Virtualization Error Detection/Isolation AMD-Vi (IOMMU 1.26) HyperTransport error handling, PCIe® Advanced Error Reporting, PCIe® end-to-end Cycle > Redundancy Check AMD-Vi (IOMMU 1.26) HyperTransport error handling, PCIe® Advanced Error Reporting, PCIe® end-to-end Cycle > Redundancy Check 15.4W/ 5.75W TSMC 65nm 29 x 29mm FCBGA AMD-Vi (IOMMU 1.26) HyperTransport error handling, PCIe® Advanced Error Reporting, PCIe® end-to-end Cycle > Redundancy Check 18W/6.15W TSMC 65nm 29 x 29mm FCBGA 12.6W/ 5.4W TSMC 65nm Package 29 x 29mm FCBGA * Using HyperTransport™ technology ** ACP stands for Average CPU power. See www.amd.com/ACP ©2012 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, AMD Opteron, AMD Virtualization, AMD-V and combinations thereof are trademarks of Advanced Micro Devices, Inc. HyperTransport is a licensed trademark of the HyperTransport Technology Consortium. Other names are for reference only and may be the trademarks of their respective owners. PID 50368F USB Ports 12 USB 2.0 + 2 USB 1.1 PCI Bus Support PCI rev 2.3 Serial ATA AHCI 1.1 SATA 3.0Gb/s with SW RAID Support SATA Ports 6 (can be independently > disabled) Max TDP/Idle 4W/1W Process Technology TSMC .13um Package 528 ball FCBGA, 21x21mm, 0.8mm pitch