AMD Investor Presentation Deck

61 Endnotes Footnotes MLN-016B, MLN-086B, GD-183, ROM-169, ROM-557, M1200-01 MLN-016B: Results as of 07/06/2021 using SPECrate®2017_int_base. The AMD EPYC 7763 scored 854, http://spec.org/cpu2017/results/res2021q3/cpu 2017-20210622-27664.html which is higher than all other 2P scores published on the SPEC® website. SPEC®, SPECrate® and SPEC CPU® are registered trademarks of the Standard Performance Evaluation Corporation. See www.spec.org for more information.. MLN-086B: SPECrate®2017_fp_base comparison based on best performing systems published at www.spec.org as of 07/06/2021. Configurations: 2x AMD EPYC 7763 (651 SPECrate®2017_fp_base, http://spec.org/cpu2017/results/res 2021q1/cpu 2017-20210219-24944.html, $15780 1Ku price total, 560W total TDP) versus 2x Intel Xeon Platinum 8380 (489 SPECrate®2017_fp_base, http://spec.org/cpu 2017/results/res2021q2/cpu 2017-20210521-26361.html, $16198 1Ku price total, 540W total TDP) for 1.33x the performance at 1.37x the score per total CPU $; 0.83x the performance/Core; 1.28x the performance/Watt. Top EPYC 7002 Series result: 2x AMD EPYC 7H12 (543 SPECrate®2017_fp_base, http://spec.org/cpu 2017/results/res2020q3/cpu2017-20200707-23397.html). AMD 1Ku pricing and Intel ARK.intel.com specifications and pricing as of 4/6/21. SPEC®, SPEC CPU®, and SPECrateⓇ are registered trademarks of the Standard Performance Evaluation Corporation. see www.spec.org for more information. GD-183: AMD Infinity Guard features vary by EPYC TM Processor generations. Infinity Guard security features must be enabled by server OEMs and/or Cloud Service Providers to operate. Check with your OEM or provider to confirm support of these features. Learn more about Infinity Guard at https://www.amd.com/en/technologies/infinity-guard. GD-183 ROM-169: For a complete list of world records see http://amd.com/world records. - ROM-557: Estimates based on AMD Server Virtualization TCO (total cost of ownership) Estimator tool v5.5, comparing the AMD EPYC TM and Intel® Xeon® server solutions required to deliver 320 total virtual machines (VM), requiring 1 core and 8GB of memory per VM, with a minimum total solution memory requirement of 2.56 TB of memory. The analysis includes both hardware and virtualization software components. For 320 VMs and 1 core per VM, the Intel _Gold_6250 processor requires 20 - 2P servers. The AMD EPYC_7702P solution requires 5 1P servers. Virtualization software pricing as of October 2019. Third party names are for informational purposes only and may be trademarks of their respective owners. This scenario contains many assumptions and estimates and, while based on AMD internal research and best approximations, should be considered an example for information purposes only, and not used as a basis for decision making over actual testing. All pricing is in USD. ROM-557 M1200-01 World's fastest data center GPU is the AMD Instinct TM MI250X. Calculations conducted by AMD Performance Labs as of Sep 15, 2021, for the AMD Instinct TM M1250X (128GB HBM2e OAM module) accelerator at 1,700 MHz peak boost engine clock resulted in 95.7 TFLOPS peak theoretical double precision (FP64 Matrix), 47.9 TFLOPS peak theoretical double precision (FP64), 95.7 TFLOPS peak theoretical single precision matrix (FP32 Matrix), 47.9 TFLOPS peak theoretical single precision (FP32), 383.0 TFLOPS peak theoretical half precision (FP16), and 383.0 TFLOPS peak theoretical Bfloat16 form at precision (BF16) floating-point performance. Calculations conducted by AMD Performance Labs as of Sep 18, 2020 for the AMD Instinct TM MI100 (32GB HBM2 PCIe® card) accelerator at 1,502 MHz peak boost engine clock resulted in 11.54 TFLOPS peak theoretical double precision (FP64), 46.1 TFLOPS peak theoretical single precision matrix (FP32), 23.1 TFLOPS peak theoretical single precision (FP32), 184.6 TFLOPS peak theoretical half precision (FP16) floating-point performance. Published results on the NVidia Ampere A100 (80GB) GPU accelerator, boost engine clock of 1410 MHz, resulted in 19.5 TFLOPS peak double precision tensor cores (FP64 Tensor Core), 9.7 TFLOPS peak double precision (FP64). 19.5 TFLOPS peak single precision (FP32), 78 TFLOPS peak half precision (FP16), 312 TFLOPS peak half precision (FP16 Tensor Flow), 39 TFLOPS peak Bfloat 16 (BF16), 312 TFLOPS peak Bfloat16 format precision (BF16 Tensor Flow), theoretical floating-point performance. The TF32 data format is not IEEE compliant and not included in this comparison. https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/nvidia-ampere-architecture-whitepaper.pdf, page 15, Table 1. AMD CORPORATE PRESENTATION | MAY 2022 AMD

View entire presentation