AMD AI Chip Unveiling - Assembly - Salesforce Research
AMD Unveils Instinct MI300 APUs In MI300A & MI300X Flavors: CDNA 3 GPU, Up To 24 Zen 4 Cores, 192 GB HBM3, 153 Billion Transistors
wccftech.com - 11 months ago - Read On Original Website
AMD has officially unveiled its Instinct MI300 APUs which combine Zen 4 CPU cores with CDNA 3 GPU cores with up to 153 Billion transistors & 192 GB HBM3 memory.
More Context
keyboard_arrow_down keyboard_arrow_right What is AMD's latest chip that has been unveiled?
videocardz.com Instinct MI300X GPU
servethehome.com Instinct MI300
wepc.com MI300X
crn.com Instinct MI300X, EPYC 97X4
finance.yahoo.com AI superchip
neowin.net 128-core EPYC 97X4 Series
insidehpc.com 4th Generation EPYC
latestly.com EPYC 97X4
techradar.com 144-Core EPYC Bergamo
wccftech.com Instinct MI300 APUs
pcmag.com Instinct MI300X
seekingalpha.com MI300 series
AMD Leads The Industry In Offering The First Exascale Instinct MI300 APU: Up To 24 Zen 4 Cores, GPU-Only MI300X With Up To 192 GB HBM3 Memory
More Context
keyboard_arrow_down keyboard_arrow_right What are the technical specifications of the chip?
anandtech.com from 16 cores to 96 cores
neowin.net 128-core
wccftech.com 6 XCDs (Up To 228 CUs), 3 CCDs (Up To 24 Zen 4 Cores), 8 HBM3 Stacks
servethehome.com 128GB of HBM3, 5.2TB/s of memory bandwidth 896GB/s of Infinity Fabric bandwidth
crn.com higher core design and greater energy efficiency
zdnet.com multiple GPU "chiplets" plus 192 gigabytes of HBM3 DRAM memory, and 5.2 terabytes per second of memory bandwidth
finance.yahoo.com it can use up to 192GB of memory
videocardz.com up to 128 cores, deliver up to 3.7x throughput performance for key cloud native workloads compared to Ampere1
prnewswire.com unmatched core count in 1U and 2U density
All three chipmakers, AMD, Intel, and NVIDIA, are currently in the race to offer the world's first CPU and CPU combo. Intel's offering was going to be Falcon Shores which was going to combine an x86 CPU with a leading-edge GPU but those plans didn't come to fruition and the Chipzilla has delayed the project for the foreseeable future.
More Context
keyboard_arrow_down keyboard_arrow_right Who are the potential competitors for AMD's chip?
crn.com Amazon Web Services, Microsoft Azure, Google Cloud and others
benzinga.com NVIDIA Corp. NVDA, Meta Platforms Inc. META, and Tesla Inc. TSLA
seekingalpha.com Nvidia Corporation
fool.com Nvidia's H100
Meanwhile, NVIDIA's answer comes in the form of the Grace and Hopper Superchips which combine the Grace ARM CPU package with the Hopper H200 GPU on the same board but that's still a far cry from AMD's chiplet-based and on-package integration of Zen 4 CPU and CDNA 3 GPU cores.
More Context
keyboard_arrow_down keyboard_arrow_right How does the chip compare to Nvidia's chips?
marketwatch.com closest competitor
zdnet.com 2.4 times the memory density
crn.com fewer GPUs
wepc.com the MI300X has up to 2.4X the HBM density and 1.6X the HBM bandwidth
finance.yahoo.com a purported rival
So as of right now, AMD is truly taking charge of offering the first integrated x86 CPU and GPU package in the form of the Instinct MI300 APU. The same APU is going to be the lead product to secure AMD's strategic dominance within the AI segment. The AMD Instinct MI300 APU is going to power the El Capitan Supercomputer, delivering over 2 Exaflops of double precision horsepower. Rounding up some of the highlighted features of the AMD Instinct MI300 APUs, we have:
More Context
keyboard_arrow_down keyboard_arrow_right What are the businesses that AMD is targeting with the new chip?
crn.com AI, Cloud
techradar.com data center & AI
finance.yahoo.com the latest stock market news and
videocardz.com cloud native and technical computing
latestly.com data centre
wepc.com high-performance computing (HPC) and AI workloads
wccftech.com CPU / GPU workloads
servethehome.com PyTorch, Hugging Face
neowin.net and for cloud data centers
cnbc.com developers and server makers
benzinga.com health care to 5G networks and data centers
zdnet.com artificial intelligence computing
beststocks.com high-performance computing, graphics, and visualization technologies
channelnewsasia.com cloud computing providers and other large chip buyers
seekingalpha.com supercomputers and traditional high-performance computing
prnewswire.com growing cloud native environments
Up To 8 Chiplets + 8 Memory Stacks (5nm + 6nm process)
MI300A Sampling Now, MI300X Sampling Next Quarter, Ramp In Q4 2023
AMD will be utilizing both 5nm and 6nm process nodes for its Instinct MI300 'CDNA 3' APUs. The chip will be outfitted with the next generation of Infinity Cache and feature the 4th Gen Infinity architecture which enables CXL 3.0 ecosystem support.
The Instinct MI300 accelerator will rock a unified memory APU architecture and new Math Formats, allowing for a 5x performance per watt uplift over CDNA 2 which is massive. AMD is also projecting over 8x the AI performance versus the CDNA 2-based Instinct MI250X accelerators. The CDNA 3 GPU's UMAA will connect the CPU and GPU to a unified HBM memory package, eliminating redundant memory copies while delivering low TCO.
The AMD Instinct MI300 will be available in two flavors, the MI300A, and the MI300X. The APU will encompass several 3D chiplet packages, all combining to house an insane 153 Billion transistors. Those transistors include various core IPs, memory interfaces, interconnects, and much more.
More Context
keyboard_arrow_down keyboard_arrow_right What are the other applications that the chip could be used for?
crn.com AI, Cloud Expansion
neowin.net business and cloud
videocardz.com cloud native and technical computing
servethehome.com NICs, storage, and even memory
wepc.com high-performance computing (HPC) and AI workloads
finance.yahoo.com large language models and generative AI
latestly.com Software Enablement for Generative AI (Artificial Intelligence)
benzinga.com health care to 5G networks and data centers
zdnet.com large language models
beststocks.com high-performance computing, graphics, and visualization technologies
prnewswire.com technical computing
The CDNA 3 architecture is the fundamental DNA of the Instinct MI300 but the APU also comes with up to 304 CDNA 4 Compute Units for a total of 24,576 cores, up to 24 Zen 4, and up to 192 GB of the next-generation HBM3 memory running in 8192-bit wide bus config. Following is how these configurations are divided in terms of GPU CUs, CPU Cores, and memory:
MI300A - 6 XCDs (Up To 228 CUs), 3 CCDs (Up To 24 Zen 4 Cores), 8 HBM3 Stacks (128 GB)
As you can see, AMD's Instinct MI300 APUs will come in various configurations with CPU-only, GPU-only, and CPU+GPU integrated SKUs. The Instinct MI300A and Instinct MI300X will target the CPU / GPU workloads.
AMD's Instinct MI300 APU lineup will be compatible with a brand new socket known as the SH5, allowing for up to four such chips to be configured together using the latest Infinity Fabric inter-connect and new interconnect switches for faster bandwidth and inter-connectivity speeds.
In the latest performance comparisons, AMD showcased that the Instinct Mi300 delivers an 8x boost in AI performance (TFLOPs) and a 5x AI performance per watt (TFLOPs/watt) boost over the Instinct MI250X. AMD's Instinct MI300A APU accelerators are sampling now and the MI300X will be sampling in Q3 2023. Both products are also expected to ramp up in Q4 of 2023.