For a specific need, wanted to increase the hugepagesize using the below commands, but it has been denied. Is there any alternative approach to increase the hugepagesize
bash -c 'echo 4 > /sys/devices/system/node/node0/hugepages/hugepages-1048576kB/nr_hugepages'
bash -c 'echo 4 > /sys/devices/system/node/node1/hugepages/hugepages-1048576kB/nr_hugepages'
bash -c 'echo 4 > /sys/kernel/mm/hugepages/hugepages-1048576kB/nr_hugepages'
The current status shows:
cat /proc/meminfo|grep Huge
AnonHugePages: 0 kB
ShmemHugePages: 0 kB
FileHugePages: 0 kB
HugePages_Total: 0 <== not available
Hugepagesize: 2048 kB <== need to increase to 1GB
Hugetlb: 0 kB
Sorry to inform you that, since DevCloud is set up uniformly as a cluster for various operations we will not be able to increase hugepage size.
Also note that, hugepaging is a complex and, if used in the wrong way, it might decrease overall performance. Example if a hugepage is swapped, reading it is only faster if one needs to access almost the entire hugepage. That means if user randomly access different parts of the memory and just read a couple of kilobytes, user should just use plain old normal pages.
We assume that your issue is resolved. If you need any additional information, please post a new question as this thread will no longer be monitored by Intel.