RISC-V
Central Processing Unit
POWER9
Server
ARM Architecture
Qualcomm Inc.
Graphics Processing Unit
System on a Chip
Western Digital
Machine Learning
NVIDIA
Intel
Instruction Set Architecture
64-Bit Computing
Google
FLOPS
Integrated Circuit
Xeon Phi
Reduced Instruction Set Computing
Supercomputer
Cloud Computing
Artificial Intelligence
Code Name
7-Nanometre
GeForce 10 Series
Multi-Core Processor
Microprocessor
Field-Programmable Gate Array
IBM
PCI Express
NVLink
Raspberry Pi
University of California, Berkeley
Hardware Acceleration
Ampere
32-Bit
SIMD
Thread
Computer Data Storage
Plug-In
Encryption
AnandTech
Digital Signal Processor
Cavium Networks
Microsoft Surface
CMOS
Open Compute Project
Apple A11.1
Cache Coherence
Data Center
Single-Precision Floating-Point Format
Floating-Point Unit
Marvell Technology Group
David A. Patterson
Address Space
Xeon
Scalability
X86
Data Storage
Free and Open-Source Software
Exascale Computing
Nvidia Tesla
Half-Precision Floating-Point Format
Floating Point
Big Data
Deep Learning
POWER8
Moore's Law
Computer Science
Laptop
Computing
Bandwidth
Computer
Single-Board Computer
The Register
Open-Source Software
Semiconductor Device Fabrication
Volta
High Bandwidth Memory
Hanoi
Microarchitecture
Data
Rackspace
Hertz
Motherboard
Domain-Specific Language
Tensor Processing Unit
Double-Precision Floating-Point Format
Knight's Cove
Calxeda
Stack
Computer Architecture
Source Code
Information Superhighway
Computer Hardware
Thermal Design Power
Computing Platform
IBM Current
Veep
Application-Specific Integrated Circuit
Interface
Ice Age
Programmer
Vector Graphic
Sunny Cove
CPU Cache
Gigabyte
Backward Compatibility
IBM POWER
Software
Meltdown
Desktop Computer
Software Development
Spectre
AI Accelerator
Mobile Device
Benchmark
Intel Core
Mali
VIA Nano
Intellectual Property
Computer Performance
Booting
File Allocation Table
DisplayPort
Assembly Language
Dual In-Line Package

Hardware for Deep Learning. Part 2: CPU - Intento

blog.inten.to
10 min read
standard
CPUs are Central Processing Units, the ordinary processors we got used to. They are typically multi-core even on the desktop market (usually from 2 to 10 cores in modern Core i3-i9 Intel CPUs, but up…
There are many manuals on how to assemble your own DL machine, find one of them if you want more details on particular hardware choices.

Intel's some recent plans on upcoming CPU platforms are leaked. They are interesting, but do not look like a game-changer regarding ML/DL. Nevertheless Intel plays in almost all other interesting niches, and we'll return to it several times. Starting right now :)

Xeon Phi

UPD: Seems to be dead now. So the text below is for history:

There are still attempts to make a heavily multi-core processors like Intel Xeon Phi with up to 72 cores. The one of the most powerful existing Phi processors, the 7290F is a 72-core (288-thread, 4 threads per core!) processor with peak performance of 3456 GFLOPS DP=FP64 (so probably 2x3456 SP=FP32 GFLOPS) (and $3368.00 recommended price, with 260W TDP) which is roughly comparable to NVIDIA GTX 1060 (1070 Ti if the FP32–64 calculations were correct) by peak performance, but not the price and power ($350–500 depending on memory size 3 or 6 Gb, 120W TDP for 1060; $800/180W for 1070 Ti). Nonetheless to meaningfully compare these different processors we have to do a benchmark on real-life ML/DL tasks.

Knights Hill was cancelled, and Intel targets at exascale computing. "One step we're taking is to replace one of the future Intel Xeon Phi processors (code name Knights Hill) with a new platform and new micro-architecture specifically designed for exascale," Intel's Trish Damkroger, a data center group veep, said.

Intel announced a Knights Mill series specialized in deep learning and there is a 72-core 7295 processor. It is using LGA3647 socket, and there are no more PCI Express versions. The performance and prices are still unknown.

It seems that Xeon Phi line will be succeeded by a family of chips codenamed Knights Cove. These will have 38 or 44 cores each, 32GB of integrated HBM2 memory, and will be based on Ice Lake Scalable Xeons due to arrive in 2019 or 2020. The 44-core part may well be two…
Grigory Sapunov
Read full article