Intel Linux Kernel Optimizations Show Huge Benefit For High Core Count Servers

Written by Michael Larabel in Software on 29 March 2023 at 01:00 PM EDT. Page 5 of 5. 19 Comments.
Stress-NG benchmark with settings of Test: Poll. Clear Linux: 240 Threads was the fastest.
Stress-NG benchmark with settings of Test: Malloc. Clear Linux: 240 Threads was the fastest.
Stress-NG benchmark with settings of Test: Semaphores. Clear Linux: 240 Threads was the fastest.
Stress-NG benchmark with settings of Test: Context Switching. Clear Linux: 240 Threads was the fastest.

The results basically speak for themselves for the benefit of Intel's work on better optimizing the Linux kernel for higher CPU core counts.

RocksDB benchmark with settings of Test: Random Fill. Clear Linux: 60 Threads was the fastest.
RocksDB benchmark with settings of Test: Random Read. Clear Linux: 120 Threads was the fastest.
RocksDB benchmark with settings of Test: Update Random. Clear Linux: 60 Threads was the fastest.
RocksDB benchmark with settings of Test: Sequential Fill. Clear Linux: 60 Threads was the fastest.
RocksDB benchmark with settings of Test: Random Fill Sync. Clear Linux: 60 Threads was the fastest.
RocksDB benchmark with settings of Test: Read Random Write Random. Clear Linux: 60 Threads was the fastest.

Facebook/Meta's RocksDB was another software package benefiting from Intel's kernel optimizations to reduce contention and other enhancements primarily for higher core counts.

Geometric Mean Of All Test Results benchmark with settings of Result Composite, Clear Linux, Ubuntu SPR Scaling. Clear Linux was the fastest.

If taking the geometric mean of all the benchmarks carried out across the different thread configurations, here is the geometric mean across the many different workloads tested. As the core/thread counts increased, Intel's Clear Linux lead greatly expanded compared to where Ubuntu 23.04 was running on this Sapphire Rapids server with its default kernel.

For many of the workloads the benefits were quite profound and hopefully the Intel engineers involved will be able to get all of their relevant optimization patches upstreamed into the mainline kernel soon. Intel's work clearly shows the Linux kernel stands to benefit from greater scalability with today's increasing core count systems, especially with AMD Bergamo and Intel Sierra Forest around the corner but already these patches can be very significant even for today's servers.

Coming up next week will be similar tests on AMD 4th Gen EPYC Genoa for showing how Intel's kernel patches are still very beneficial there too and showing how RHEL9 fits into the equation too.

If you enjoyed this article consider joining Phoronix Premium to view this site ad-free, multi-page articles on a single page, and other benefits. PayPal or Stripe tips are also graciously accepted. Thanks for your support.


Related Articles
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.