Announcement

Collapse
No announcement yet.

NVIDIA Engineer Shows Off New Linux Patches For Proactive Memory Compaction

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • NVIDIA Engineer Shows Off New Linux Patches For Proactive Memory Compaction

    Phoronix: NVIDIA Engineer Shows Off New Linux Patches For Proactive Memory Compaction

    Veteran Linux engineer Nitin Gupta of NVIDIA has unveiled his latest patches on the work he got started on last year: proactive memory compaction for Linux motivated by the latency issues brought on by he current on-demand compaction when an application requests a lot of hugepages...

    http://www.phoronix.com/scan.php?pag...RAM-Compaction

  • #2
    Relevant issue: https://github.com/OpenNebula/one/issues/3124

    Comment


    • #3
      Which kernel will integrate this patch?

      Comment


      • #4
        Originally posted by Azrael5 View Post
        Which kernel will integrate this patch?
        Unknown, this feature is still under development.

        Comment


        • #5
          I've been thinking about KSM for more than just KVM/madvise.
          Think of the benefits to those running 1 instance of the same app on each thread.
          Something like transparent huge page merging that don't need apps to specifically call functions to take advantage of it.

          compaction + merging would be like doubling or even 16x your mem in specific scenarios. The cache benefits would be potentially massive too.

          Comment


          • #6
            Check out my blog post on Proactive Compaction to learn more: https://nitingupta.dev/post/proactive-compaction/

            Comment

            Working...
            X