 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
I calculate the Max Flops on Skylake with cpu frequency*16.
Is cpufrequency*32 on GOLD version of Skylake?
Link Copied
 « Previous

 1
 2
 Next »
 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
The "Base AVX512 Core Frequency (GHz)" is the frequency that the processor will use when running 512bit SIMD instructions in the absence of any Turbo boost. The columns to the right show the maximum frequency that the processor will use when running 512bit SIMD instructions with various numbers of active cores.
The maximum frequency is only available if the temperature does not exceed the temperature limits, the package power does not exceed the package power limits, and the electrical current does not exceed the current limits. If any of these limits are exceeded, the frequencies will be reduced until none of the limits are exceeded.
The "Base AVX512 Core Frequency (GHz)" is also intended to represent the minimum frequency that will ever be seen for any powerlimited workload (assuming a correctly configured cooling system). So this frequency can be used to compute a lower bound on the peak GFLOPS. For example, the Xeon Platinum 8180 as a "Base AVX512 Core Frequency" of 1.7 GHz, with 28 cores, and two AVX512 units per core, giving a lower bound
Lower Bound: 28 cores * 1.7 GHz * 32 FLOPS/Hz = 1523.2 GFLOPS (per socket).
The maximum 28core AVX512 frequency of 2.3 GHz provides an upper bound on the peak performance
Upper Bound: 28 cores * 2.3 GHz * 32 FLOPS/Hz = 2060.8 GFLOPS (per socket)
The actual frequency when running computeintensive AVX512 workloads depends on the unique characteristics of the specific piece of silicon (particularly leakage current), as well as the characteristics of the cooling system (ambient temperature, heat sink thermal conductivity, air flow rate, etc).
We have 3472 Xeon Platinum 8160 (24core0 processors in 1736 twosocket nodes. The Base AVX512 Core Frequency for these processors is 1.4 GHz and the maximum 24core AVX512 frequency is 2.0 GHz. When running Intel's optimized LINPACK benchmark, we see that the average frequency of these processors varies between about 1.52 GHz and about 1.73 GHz, with sustained (LINPACK) performance varying by the same proportions.
 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
For example, the Xeon Platinum 8180, Processor Base Frequency is 2.5GHz, Max Turbo Frequency is 3.8GHz(from ark.intel.com). May I use 2.5 or 3.8 to calculate the Max GFLOPS?
 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
You should use 2.5 to calculate Max Gflops. You can't achieve the Max Turbo if you are using all the cores.
 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
The rightmost column of the first row of the table shows that 2.3 GHz is the maximum Turbo frequency that the Xeon Platinum 8180 allows when using all cores and running AVX512 code. 2.3*32*28 = 2060.8 GFLOPS for doubleprecision FMA operations on 512bit vectors.
 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
CPU processor Max Turbo Frequency is 3.8GHz, but AVX512 Max Turbo Frequency is only 2.3GHz. Why is this?
And calculate Max GFLOPS always use ( CPUBaseFrequency * cores * 32(or 16, 8, 4) ) before.
 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
Are there any relations between AVX512 Frequency and CPU Processor Frequency?
 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
The nomenclature can be confusing. The "nonAVX" frequencies, the "AVX 2.0" frequencies, and the "AVX512" frequencies are all CPU core frequencies, with the different tables applying to code with different SIMD register widths in use.
For any CPU model, the maximum frequency that each core can run at is determined by both the number of active cores and by the width of the SIMD units that are active in each core. In figure 3 above, the label "AVX512 Turbo Frequencies" means the maximum frequency that any core can run at when the 512bit SIMD units are active. (The 512bit SIMD units are active whenever the processor has executed an instruction using 512bit registers in the last millisecond.)
As an example, we can compare the values in Figures 1, 2, 3 from the "Intel Xeon Processor Family Specification Update" (document 336065005, February 2018) for the Xeon Platinum 8180. If there are 20 active cores:
 any of the cores with the 512bit SIMD units active can run at up to 2.6 GHz (Figure 3, row 1, column for 20 cores active)
 any of the cores with the 256bit SIMD units active can run at up to 3.1 GHz (Figure 2, row 1, column for 20 cores active)
 any of the cores with neither 256bit nor 512bit SIMD units active can run at up to 3.5 GHz (Figure 1, row 1, column for 20 cores active)
These tables provide the maximum turbo frequencies for each case  the actual frequencies will be reduced if needed to keep the running average power under the TDP limit for the processor model.
 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
Hi McCalpin, your comments is really very helpful. But I still have a question and I didn't find out the answer from Google or Intel Official web site.
1.We usually calculate CPU performance by Frequency*IPC*Cores. But Intel does not seem to release IPC of each CPU. So my first question is how to get IPC of each CPU.
2.According to your comments above, it seems we can not use the processor basic frequency released by Intel on the official site. Is there a formula which can calculate the CPU performance using the Processor Basic Frequency ?
Thank you very much.
McCalpin, John (Blackbelt) wrote:The nomenclature can be confusing. The "nonAVX" frequencies, the "AVX 2.0" frequencies, and the "AVX512" frequencies are all CPU core frequencies, with the different tables applying to code with different SIMD register widths in use.
For any CPU model, the maximum frequency that each core can run at is determined by both the number of active cores and by the width of the SIMD units that are active in each core. In figure 3 above, the label "AVX512 Turbo Frequencies" means the maximum frequency that any core can run at when the 512bit SIMD units are active. (The 512bit SIMD units are active whenever the processor has executed an instruction using 512bit registers in the last millisecond.)
As an example, we can compare the values in Figures 1, 2, 3 from the "Intel Xeon Processor Family Specification Update" (document 336065005, February 2018) for the Xeon Platinum 8180. If there are 20 active cores:
any of the cores with the 512bit SIMD units active can run at up to 2.6 GHz (Figure 3, row 1, column for 20 cores active)
any of the cores with the 256bit SIMD units active can run at up to 3.1 GHz (Figure 2, row 1, column for 20 cores active)
any of the cores with neither 256bit nor 512bit SIMD units active can run at up to 3.5 GHz (Figure 1, row 1, column for 20 cores active)
These tables provide the maximum turbo frequencies for each case  the actual frequencies will be reduced if needed to keep the running average power under the TDP limit for the processor model.
 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
(1) The information you need on max operations per cycle is not easy to find, but all the available information is contained in the posts above.
(2) There is no formula to determine the maximum allcore Turbo frequency from the "nominal" frequency. You have to look up the modelspecific information in the "Specification Update" document for the processor family you are interested in.
 Mark as New
 Bookmark
 Subscribe
 Mute
 Subscribe to RSS Feed
 Permalink
 Report Inappropriate Content
Thank you very much. I have already got the information you mentioned.
McCalpin, John (Blackbelt) wrote:(1) The information you need on max operations per cycle is not easy to find, but all the available information is contained in the posts above.
(2) There is no formula to determine the maximum allcore Turbo frequency from the "nominal" frequency. You have to look up the modelspecific information in the "Specification Update" document for the processor family you are interested in.
 Subscribe to RSS Feed
 Mark Topic as New
 Mark Topic as Read
 Float this Topic for Current User
 Bookmark
 Subscribe
 Printer Friendly Page
 « Previous

 1
 2
 Next »