Data centres and servers
Xeon, EPYC, Graviton, H100 and the chips of the cloud. 296 chips on EveryChip so far.
Data centres and cloud servers use the largest and most power-hungry chips: server CPUs with up to a few hundred cores, AI accelerators drawing 700 W to over 1,000 W each, network chips and DPUs, and huge amounts of memory. Intel's Xeon and AMD's Epyc dominate general-purpose servers, while Arm designs such as Amazon's Graviton have grown fast. Nvidia's GPUs lead in AI.
Server CPUs
A server processor is built for many tasks at once and for long, reliable running. It supports large amounts of error-correcting memory, many PCIe lanes for storage and network cards, and two or more sockets on one board. AMD's Epyc, launched in 2017, used chiplets to fit more cores than single-die designs, and took a large share of the market from Intel's Xeon.
Custom cloud chips
The largest cloud companies now design their own chips. Amazon has Graviton CPUs and Trainium and Inferentia AI chips, Google its TPU and Axion CPU, and Microsoft its Cobalt CPU and Maia accelerator. They buy fewer standard parts this way and tune chips to their own software. The power used by AI servers has become a constraint on where data centres can be built.
Famous data centres and servers chips
Still to be written: AWS Graviton.
Further reading
🇺🇸 United StatesIvy Bridge
Intel's 22 nm tick microarchitecture featuring FinFET transistors and integrated graphics.CPU · x86 · 2012
🇺🇸 United StatesZen
AMD's processor microarchitecture family behind Ryzen, Threadripper and Epyc chipsCPU · x86 · 2017
🇺🇸 United StatesXeon D
Intel's low-power x86 system-on-chip line for microserversCPU · x86 · 2015
🇺🇸 United StatesNetBurst
Intel's Pentium 4-era x86 CPU design, built for very high clock speedsCPU · x86 · 2000
Photo credits on this page
The small pictures above come from Wikimedia Commons under free licences. Some are cropped to fit.
- IBM Telum: IBM Research, CC BY 3.0, source
- AMD Instinct: Мой Компьютер, CC BY 3.0, source
- DDR5 SDRAM: Jacek Halicki, CC BY-SA 4.0, source
- Power10: Henriok, CC0, source
- Sunny Cove: Fritzchens Fritz, CC BY 2.0, source
- Xeon: Rjluna2, CC BY-SA 4.0, source
- Wolfdale: Kyro, Public domain, source
- PowerPC 600: Konstantin Lanzet, CC BY-SA 3.0, source
- Pentium Pro: Thomas Nguyen, CC BY-SA 4.0, source
- POWER9: IBM España, Public domain, source
- Transputer: NobbiP, CC BY-SA 3.0, source
- Ivy Bridge: IrisChronomia, CC BY-SA 4.0, source
- Ryzen Threadripper: Geni, CC BY-SA 4.0, source
- Epyc: Smial ( talk ), FAL, source
- Zen 3: Fritzchens Fritz, CC BY 2.0, source
- AMD K8: Redfox, Public domain, source
- Piledriver: D-Kuru, CC BY-SA 4.0, source
- Granite Rapids: Intel Corporation, Public domain, source
- CVAX: wdwd, CC BY-SA 3.0, source
- Itanium: Konstantin Lanzet (with permission), CC BY-SA 3.0, source
- NetBurst: mitopencourseware, CC BY-SA 4.0, source
- P6: Fritzchens Fritz, CC0, source
- Xeon D: Intel Corporation, Public domain, source
- Kaby Lake: Eric Gaba ( Sting - fr:Sting ), CC BY-SA 4.0, source
- Coffee Lake: ElooKoN, CC BY-SA 4.0, source
- Raptor Lake: Fritzchens Fritz, CC0, source
- Comet Lake: Smial ( talk ), FAL, source
- Turing: Fritzchens Fritz, CC0, source
- Radeon Pro: David290, CC BY-SA 4.0, source
- DDR2 SDRAM: An-d, CC BY-SA 3.0, source
- MC88100: Pauli Rautakorpi, CC BY 3.0, source
- Silvermont: Raimond Spekking, CC BY-SA 4.0, source
- PowerPC 970: Henriok, CC0, source
- Haswell: Intel Free Press, CC BY 2.0, source
- Tensor Processing Unit: Norman P. Jouppi, George Kurian, Sheng Li, Peter Ma, Rahul Nagarajan,, CC BY 4.0, source
- Hopper: 极客湾Geekerwan, CC BY 3.0, source