Non-Uniform Memory Access or Non-Uniform Memory Architecture (NUMA) is a physical memory design used in SMP (multiprocessors) architecture, where the memory access time depends on the memory location relative to a processor. Under NUMA, a processor can access its own local memory faster than non-local memory, that is, memory local to another processor or memory shared between processors.
In a NUMA system, processors, memory, and I/O are grouped together into cells. The latency and bandwidth characteristics of communication within a cell are “fast” while going outside a cell is “slow”. Since the memory in NUMA systems is physically distributed but logically shared, these systems offer better performance to applications that are optimized to use their features. For non-optimized applications, they still offer better performance since the default behaviour is designed to be benign if not beneficial and they still have access to much larger shared resources of memory, CPUs, and disk space.
How can we identify if NUMA is enabled
NUMA has to enabled in the BIOS. If dmesg does not have records of Numa initialization during bootup, then it is possible that NUMA related messages in the kernel ring buffer might have been overwritten.
NUMA Enabled Systems
If NUMA is enabled on BIOS, then execute the command ‘numactl –hardware‘ to list inventory of available nodes on the system. Below is example output of numactl –hardware on a system which has NUMA.
# numactl --hardware available: 2 nodes (0-1) node 0 cpus: 0 1 2 3 4 5 6 7 16 17 18 19 20 21 22 23 node 0 size: 8157 MB node 0 free: 88 MB node 1 cpus: 8 9 10 11 12 13 14 15 24 25 26 27 28 29 30 31 node 1 size: 8191 MB node 1 free: 5176 MB node distances: node 0 1 0: 10 20 1: 20 10
NUMA Disabled Systems
1. If NUMA is disabled on BIOS, then the command ‘numactl –show‘ does not show multiple nodes.
# numactl --show policy: default preferred node: current physcpubind: 0 1 2 3 4 5 6 7 8 9 10 11 cpubind: 0 nodebind: 0 membind: 0
2. The command ‘numactl –hardware‘ also does not list multiple nodes.
available: 1 nodes (0) node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 node 0 size: 65525 MB node 0 free: 17419 MB node distances: node 0 0: 10
3. If the server does not have NUMA support or if the BIOS option is not enabled, then the following messages will be seen in dmesg
No NUMA configuration found Faking a node at 0000000000000000-0000001027fff000
4. If ACPI is disabled, that will also disable NUMA; verify that ACPI is not disabled by a grub.conf kernel parameter and remove it if found:
$ grep acpi=off /proc/cmdline $
