Intel's Linux Software Optimizations Continue Paying Off Big Time For Xeon Emerald Rapids

Written by Michael Larabel in Operating Systems on 21 March 2024 at 11:30 AM EDT. Page 4 of 5. 15 Comments.
Y-Cruncher benchmark with settings of Pi Digits To Calculate: 1B. Clear Linux 41150 was the fastest.
Y-Cruncher benchmark with settings of Pi Digits To Calculate: 5B. Ubuntu 23.10 was the fastest.
ClickHouse benchmark with settings of 100M Rows Hits Dataset, First Run / Cold Cache. Clear Linux 41150 was the fastest.
ClickHouse benchmark with settings of 100M Rows Hits Dataset, Second Run. Clear Linux 41150 was the fastest.
Redis benchmark with settings of Test: GET, Parallel Connections: 500. Clear Linux 41150 was the fastest.
Redis benchmark with settings of Test: SET, Parallel Connections: 500. Clear Linux 41150 was the fastest.

A number of the Clear Linux optimizations over the years have been upstreamed into the Linux kernel and other software components where relevant, but still there is more work to be done on the part of Linux distribution vendors. At least in recent months we've seen more Linux distributions considering optional or mandated x86_64-v2/v3 support or making other use of the x86_64 micro-architecture feature levels.

Graph500 benchmark with settings of Scale: 26. CentOS Stream 9 was the fastest.
Graph500 benchmark with settings of Scale: 26. Clear Linux 41150 was the fastest.
Graph500 benchmark with settings of Scale: 26. Clear Linux 41150 was the fastest.
Graph500 benchmark with settings of Scale: 26. Clear Linux 41150 was the fastest.
GROMACS benchmark with settings of Implementation: MPI CPU, Input: water_GMX50_bare. Clear Linux 41150 was the fastest.
RawTherapee benchmark with settings of Total Benchmark Time. Clear Linux 41150 was the fastest.

The performance benefits to Clear Linux in some workloads can be staggering for out-of-the-box speed on Intel Xeon Emerald Rapids.


Related Articles