Our blog

Home Our Blog

Friday07 August 2020

How to calculate peak theoretical performance of a CPU-based HPC system

Updated on November 3, 2014

To calculate peak theoretical performance of a HPC system we first need to calculate peak theoretical performance of one node (server) in GFlops and than just multiply node performance on the number of nodes your HPC system has.

HPC world is using the following formulae for node peak theoretical performance:

Node performance in GFlops = (CPU speed in GHz) x (number of CPU cores) x (CPU instruction per cycle) x (number of CPUs per node)

Where to take a number of CPU instructions per cycle:
- Intel X5600 series CPUs and AMD 6100/6200/6300 series CPUs have 4 instructions per cycle
- Intel E5-2600v1 and E5-2600v2 series CPUs have 8 instructions per cycle
Intel E5-2600v3 series CPUs have 16 instructions per cycle (as E5-2600v3 series CPUs have AVX2.0 and FMA instruction sets that at their theoretical maximum are two times lager than that of E5-2600v1 and E5-2600v2)


Example 1: Dual-CPU server based on Intel X5675 (3.06GHz 6-cores) CPUs:
3.06 x 6 x 4 x 2 = 144.88 GFLOPS

Example 2: Dual-CPU server based on Intel E5-2670 (2.6GHz 8-cores) CPUs:
2.6 x 8 x 8 x 2 = 332.8 GFLOPS
(Note that the number of instructions per cycle for E5-2600v1 and E5-2600v2 series CPUs is equal to 8)

Example 3: Dual-CPU server based on Intel E5-2690v3 (2.6GHz 12-cores) CPUs:
2.6 x 12 x 16 x 2 = 998.4 GFLOPS
(Note that the number of instructions per cycle for E5-2600v3 series CPUs is equal to 16)


JA Minisite

2857.orig.q75.o0 - Copy IntelTPP amd    qctlogo e   Mellanox APAC Partner   1