Hugepage: Difference between revisions
No edit summary |
No edit summary |
||
| Line 2: | Line 2: | ||
In Database, Huge pages is especially useful on systems like database servers. Processes like MySQL and PostgreSQL can make use of huge pages if they are enabled, and will put less strain on your RAM cache | In Database, Huge pages is especially useful on systems like database servers. Processes like MySQL and PostgreSQL can make use of huge pages if they are enabled, and will put less strain on your RAM cache<ref>https://linuxconfig.org/how-to-enable-hugepages-on-linux</ref> | ||
| Line 8: | Line 9: | ||
most standard on modern systems would be 2 MB<syntaxhighlight lang="bash"> | most standard on modern systems would be 2 MB while for 64-bit applications, it is recommended to use 1 GB hugepages if the platform supports them.<syntaxhighlight lang="bash"> | ||
$ grep -i huge /proc/meminfo | $ grep -i huge /proc/meminfo | ||
AnonHugePages: 0 kB | AnonHugePages: 0 kB | ||
| Line 20: | Line 21: | ||
Hugetlb: 0 kB | Hugetlb: 0 kB | ||
</syntaxhighlight>we can change Hugepagesize <syntaxhighlight lang="bash"> | </syntaxhighlight>we can change Hugepagesize <syntaxhighlight lang="bash"> | ||
$ sudo sysctl -w vm.nr_hugepages= | $ sudo sysctl -w vm.nr_hugepages=1024000 # for example 1G | ||
[sudo] password for hpcmate: | [sudo] password for hpcmate: | ||
vm.nr_hugepages = 102400 | vm.nr_hugepages = 102400 | ||
$ grep -i huge /proc/meminfo | |||
AnonHugePages: 0 kB | AnonHugePages: 0 kB | ||
ShmemHugePages: 0 kB | ShmemHugePages: 0 kB | ||
FileHugePages: 0 kB | FileHugePages: 0 kB | ||
HugePages_Total: | HugePages_Total: 127542 | ||
HugePages_Free: | HugePages_Free: 127542 | ||
HugePages_Rsvd: 0 | HugePages_Rsvd: 0 | ||
HugePages_Surp: 0 | HugePages_Surp: 0 | ||
Hugepagesize: 2048 kB | Hugepagesize: 2048 kB | ||
Hugetlb: | Hugetlb: 261206016 kB | ||
</syntaxhighlight>To make it persistent across futher reboot, add following line in /etc/sysctl.conf and reboot<syntaxhighlight lang="bash"> | </syntaxhighlight>To make it persistent across futher reboot, add following line in /etc/sysctl.conf and reboot<syntaxhighlight lang="bash"> | ||
vm.nr_hugepages = | vm.nr_hugepages = 1024000 | ||
</syntaxhighlight><syntaxhighlight lang="bash"> | </syntaxhighlight><syntaxhighlight lang="bash"> | ||
#Check kernel support HugeTLB | #Check kernel support HugeTLB | ||
| Line 51: | Line 52: | ||
CONFIG_HUGETLBFS=y | CONFIG_HUGETLBFS=y | ||
CONFIG_HUGETLB_PAGE=y | CONFIG_HUGETLB_PAGE=y | ||
</syntaxhighlight> | </syntaxhighlight> | ||
Some kernel versions may not allow reserving 1 GB hugepages at run time, so reserving them at boot time may be the only option by pass the hugepages option to the kernel.<syntaxhighlight lang="bash"> | Some kernel versions may not allow reserving 1 GB hugepages at run time, so reserving them at boot time may be the only option by pass the hugepages option to the kernel.<syntaxhighlight lang="bash"> | ||
default_hugepagesz=1G hugepagesz=1G hugepages=4 | default_hugepagesz=1G hugepagesz=1G hugepages=4 | ||
</syntaxhighlight> | </syntaxhighlight> | ||
== Reference == | == Reference == | ||
<references /> | <references /> | ||
Revision as of 17:22, 16 June 2023
Computer memory is allocated to processes as pages. Usually these pages are rather small (4K), meaning that a process consuming a lot of memory will also be consuming a lot of pages. Searching through a multitude of pages can result in system slow downs, which is why some servers can benefit from enabling huge pages.
In Database, Huge pages is especially useful on systems like database servers. Processes like MySQL and PostgreSQL can make use of huge pages if they are enabled, and will put less strain on your RAM cache[1]
In Network stack, Hugepage support is required for the large memory pool allocation used for packet buffers (the HUGETLBFS option must be enabled in the running kernel as indicated the previous section). By using hugepage allocations, performance is increased since fewer pages are needed, and therefore less Translation Lookaside Buffers (TLBs, high speed translation caches), which reduce the time it takes to translate a virtual page address to a physical page address. Without hugepages, high TLB miss rates would occur with the standard 4k page size, slowing performance.[2]
most standard on modern systems would be 2 MB while for 64-bit applications, it is recommended to use 1 GB hugepages if the platform supports them.
$ grep -i huge /proc/meminfo
AnonHugePages: 0 kB
ShmemHugePages: 0 kB
FileHugePages: 0 kB
HugePages_Total: 0
HugePages_Free: 0
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 2048 kB
Hugetlb: 0 kB
we can change Hugepagesize
$ sudo sysctl -w vm.nr_hugepages=1024000 # for example 1G
[sudo] password for hpcmate:
vm.nr_hugepages = 102400
$ grep -i huge /proc/meminfo
AnonHugePages: 0 kB
ShmemHugePages: 0 kB
FileHugePages: 0 kB
HugePages_Total: 127542
HugePages_Free: 127542
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 2048 kB
Hugetlb: 261206016 kB
To make it persistent across futher reboot, add following line in /etc/sysctl.conf and reboot
vm.nr_hugepages = 1024000
#Check kernel support HugeTLB
$grep -i huge /boot/config-5.4.0-150-generic
CONFIG_CGROUP_HUGETLB=y
CONFIG_ARCH_WANT_GENERAL_HUGETLB=y
CONFIG_ARCH_ENABLE_HUGEPAGE_MIGRATION=y
CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE=y
CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD=y
CONFIG_HAVE_ARCH_HUGE_VMAP=y
CONFIG_ARCH_WANT_HUGE_PMD_SHARE=y
CONFIG_TRANSPARENT_HUGEPAGE=y
# CONFIG_TRANSPARENT_HUGEPAGE_ALWAYS is not set
CONFIG_TRANSPARENT_HUGEPAGE_MADVISE=y
CONFIG_TRANSPARENT_HUGE_PAGECACHE=y
CONFIG_HUGETLBFS=y
CONFIG_HUGETLB_PAGE=y
Some kernel versions may not allow reserving 1 GB hugepages at run time, so reserving them at boot time may be the only option by pass the hugepages option to the kernel.
default_hugepagesz=1G hugepagesz=1G hugepages=4