From WikiChip
Difference between revisions of "intel/xeon phi/7295"
< intel‎ | xeon phi

Line 41: Line 41:
  
 
{{unknown features}}
 
{{unknown features}}
 +
 +
 +
== Cache ==
 +
{{main|intel/microarchitectures/knights_mill#Memory_Hierarchy|l1=Knights Mill § Cache}}
 +
{{cache size
 +
|l1 cache=4.5 MiB
 +
|l1i cache=2.25 MiB
 +
|l1i break=72x32 KiB
 +
|l1i desc=8-way set associative
 +
|l1d cache=2.25 MiB
 +
|l1d break=72x32 KiB
 +
|l1d desc=8-way set associative
 +
|l2 cache=36 MiB
 +
|l2 break=36x1 MiB
 +
|l2 desc=16-way set associative
 +
}}

Revision as of 20:00, 18 December 2017

Edit Values
Xeon Phi 7295
xeon phi knights mill (front).png
General Info
DesignerIntel
ManufacturerIntel
Model Number7295
Part NumberHJ8068303823700
S-SpecSR3VD
MarketWorkstation, Server, Supercomputer
IntroductionDecember 18, 2017 (announced)
December 18, 2017 (launched)
ShopAmazon
General Specs
FamilyXeon Phi
Seriesx205
LockedYes
Frequency1,500 MHz
Turbo Frequency1,600 MHz (1 core)
Clock multiplier15
Microarchitecture
ISAx86-64 (x86)
MicroarchitectureKnights Mill
Core SteppingA0
Process14 nm
TechnologyCMOS
Word Size64 bit
Cores72
Threads288
Max Memory384 GiB
Multiprocessing
Max SMP1-Way (Uniprocessor)
Electrical
Vcore0.55 V-1.2 V
TDP320 W
Tcase0 °C – 77 °C
Packaging
Template:packages/intel/fclga-3647

Xeon Phi 7295 is a 72-core x86 many-core microprocessor introduced by Intel in late 2017 specifically designed to accelerate artificial intelligence workloads. This processor, which is fabricated on their own 14 nm process based on the Knights Mill microarchitecture, operates at a frequency of 1.5 GHz with a TDP of 320 W and a turbo of 1.6 GHz.


DIL16 Blank.svg Preliminary Data! Information presented in this article deal with a microprocessor or chip that was recently announced or leaked, thus missing information regarding its features and exact specification. Information may be incomplete and can change by final release.


Cache

Main article: Knights Mill § Cache

[Edit/Modify Cache Info]

hierarchy icon.svg
Cache Organization
Cache is a hardware component containing a relatively small and extremely fast memory designed to speed up the performance of a CPU by preparing ahead of time the data it needs to read from a relatively slower medium such as main memory.

The organization and amount of cache can have a large impact on the performance, power consumption, die size, and consequently cost of the IC.

Cache is specified by its size, number of sets, associativity, block size, sub-block size, and fetch and write-back policies.

Note: All units are in kibibytes and mebibytes.
L1$4.5 MiB
4,608 KiB
4,718,592 B
L1I$2.25 MiB
2,304 KiB
2,359,296 B
72x32 KiB8-way set associative 
L1D$2.25 MiB
2,304 KiB
2,359,296 B
72x32 KiB8-way set associative 

L2$36 MiB
36,864 KiB
37,748,736 B
0.0352 GiB
  36x1 MiB16-way set associative 
Facts about "Xeon Phi 7295 - Intel"
base frequency1,500 MHz (1.5 GHz, 1,500,000 kHz) +
clock multiplier15 +
core count72 +
core steppingA0 +
core voltage (max)1.2 V (12 dV, 120 cV, 1,200 mV) +
core voltage (min)0.55 V (5.5 dV, 55 cV, 550 mV) +
designerIntel +
familyXeon Phi +
first announcedDecember 18, 2017 +
first launchedDecember 18, 2017 +
full page nameintel/xeon phi/7295 +
has locked clock multipliertrue +
instance ofmicroprocessor +
isax86-64 +
isa familyx86 +
l1$ size4,608 KiB (4,718,592 B, 4.5 MiB) +
l1d$ description8-way set associative +
l1d$ size2,304 KiB (2,359,296 B, 2.25 MiB) +
l1i$ description8-way set associative +
l1i$ size2,304 KiB (2,359,296 B, 2.25 MiB) +
l2$ description16-way set associative +
l2$ size36 MiB (36,864 KiB, 37,748,736 B, 0.0352 GiB) +
ldateDecember 18, 2017 +
main imageFile:xeon phi knights mill (front).png +
manufacturerIntel +
market segmentWorkstation +, Server + and Supercomputer +
max case temperature350.15 K (77 °C, 170.6 °F, 630.27 °R) +
max cpu count1 +
max memory393,216 MiB (402,653,184 KiB, 412,316,860,416 B, 384 GiB, 0.375 TiB) +
microarchitectureKnights Mill +
min case temperature273.15 K (0 °C, 32 °F, 491.67 °R) +
model number7295 +
nameXeon Phi 7295 +
part numberHJ8068303823700 +
process14 nm (0.014 μm, 1.4e-5 mm) +
s-specSR3VD +
seriesx205 +
smp max ways1 +
tdp320 W (320,000 mW, 0.429 hp, 0.32 kW) +
technologyCMOS +
thread count288 +
turbo frequency (1 core)1,600 MHz (1.6 GHz, 1,600,000 kHz) +
word size64 bit (8 octets, 16 nibbles) +