Database Performance & Variability

Results are very different with respect to transactional database benchmarks (HammerDB & OLTP). Intel's 8160 has an advantage of 22 to 29%, which is very similar to what we saw in our own independent benchmarking.

One of the main reasons is data locality: data is distributed over the many NUMA nodes causing extra latency for data access. Especially when data is locked, this can cause performance degradation.

Intel measured this with their own Memory Latency Checker (version 3.4), but you do not have rely on Intel alone. AMD reported similar results on the Linley Processor conference, and we saw similar results too.

There is more: Intel's engineers noticed quite a bit of performance variation between different runs.

Intel engineers claim that what they reported in the first graph on this page is, in fact, the best of 10 runs. Between the 10 runs, it is claimed there was a lot of variability: ignoring the outlier number 2, there are several occasions where performance was around 60% of the best reported value. Although we can not confirm that the performance of the EPYC system varies precisely that much, we have definitely seen more variation in our EPYC benchmarks than on a comparable Intel system.

Enterprise & Cloud Benchmarks HPC Benchmarks
POST A COMMENT

103 Comments

View All Comments

  • hsupengjun - Sunday, December 03, 2017 - link

    Wow, the first few pages are sooo biased, but damn, are they rightfully so. Reply
  • ajc9988 - Tuesday, January 16, 2018 - link

    @Ian Cutress & Johan De Gelas - Could you please update this by running your own numbers AFTER the full implementation of Spectre and Meltdown fixes. That would be so helpful in showing how much these have effected both platforms and whether your conclusions remain after the fixes. Thank you! Reply
  • FentonW - Wednesday, January 17, 2018 - link

    What this doesn't really address is Memory configurations.
    RAM configurations are very much limited with intel given the 2 DIMMS per channel configuration and 6 channels vs 8 with AMD.

    With Intel you can only get 384GB with 16GB DIMMS, compared to 512GB with AMD.
    If you need 512GB then you have to use 32GB DIMMS on intel which again pushes the price up considerably.
    Which is why customers often choose a Broadwell system over Skylake, to keep memory costs down.
    Reply

Log in

Don't have an account? Sign up now