Performance Optimization - Revision history

85.195.252.62: /* Debian (Ubuntu/Mint/Lubuntu/PopOS/etc) */

2021-09-10T16:56:27Z

Debian (Ubuntu/Mint/Lubuntu/PopOS/etc)

2020-03-01T22:06:57Z

Verifying CPU Affinity

2020-03-01T21:49:41Z

vcpupin

2020-03-01T21:48:13Z

vcpupin

2020-03-01T21:42:44Z

Verifying CPU Affinity

2020-03-01T21:40:33Z

Verifying CPU Affinity

2020-03-01T21:40:12Z

Verifying CPU Affinity

2020-03-01T21:38:53Z

vcpupin

2020-03-01T21:36:46Z

Verifying CPU Affinity

2020-03-01T21:35:45Z

@@ Line 6: / Line 6: @@
 <span style="color: red;">'''Note: When hugepages are configured this portion of memory is taken away from the Host. This means the Host will no longer be able to use it. Keep this in mind.'''</span>
-=== Debain (Ubuntu/Mint/Lubuntu/PopOS/etc) ===
+=== Debian (Ubuntu/Mint/Lubuntu/PopOS/etc) ===
 First check if Linux isn't already using Hugepages with: <code>cat /proc/meminfo | grep Huge</code>.
 If the output resembles the following:

@@ Line 134: / Line 134: @@
       11
 </syntaxhighlight>
-In the event that your system only uses UMA(Uniform Memory Access) the impact is definitely less pronounced but some benefit can still be gained.
+In the event that your system only uses UMA(Uniform Memory Access) the impact of pinning threads is definitely less pronounced but some benefit can still be gained.

@@ Line 96: / Line 96: @@
 vcpupin is the process where-in each vCPU assigned to the VM is tied to a physical core/thread. Configuring this has the most profound impact when dealing with a system that has multiple NUMA Nodes because it forces requests to memory to stay on one node without having to cross Intel's QPI links or AMD's Infinity Fabric. This also helps by tying the vCPUs to the Node that is directly connected to the GPU.
-<span style="color: red;">'''NOTE: In many cases a function in the BIOS known as Memory Interleave will obfuscate the NUMA Nodes making the system treat multiple sockets or multiple dies as one UMA(Uniform Memory Access) Node. This is a problem for multiple reasons. To fix this find Memory Interleave in your BIOS and set it from Auto -> Channel. If Memory Interleave does not have these options on a single socket system then chances are the system only operates in UMA mode.'''</span>
+<span style="color: red;">'''NOTE: In many cases a function in the BIOS known as Memory Interleave will obfuscate the NUMA Nodes making the system treat multiple sockets or multiple dies as one UMA(Uniform Memory Access) Node. This is a problem for multiple reasons. To fix this find Memory Interleave in your BIOS and set it from Auto -> Channel. If Memory Interleave does not have these options on a single socket system then chances are the system only operates in UMA mode. In such case this is fine.'''</span>
 === Identifying CPU Affinity ===

@@ Line 95: / Line 95: @@
 == vcpupin ==
 vcpupin is the process where-in each vCPU assigned to the VM is tied to a physical core/thread. Configuring this has the most profound impact when dealing with a system that has multiple NUMA Nodes because it forces requests to memory to stay on one node without having to cross Intel's QPI links or AMD's Infinity Fabric. This also helps by tying the vCPUs to the Node that is directly connected to the GPU.
 === Identifying CPU Affinity ===

← Older revision		Revision as of 21:42, 1 March 2020
Line 132:		Line 132:
	7 11		7 11
	</syntaxhighlight>		</syntaxhighlight>
		+	In the event that your system only uses UMA(Uniform Memory Access) the impact is definitely less pronounced but some benefit can still be gained.

@@ Line 119: / Line 119: @@
 === Verifying CPU Affinity ===
-To verify if the CPU threads have been successfully pinned to the VM the <code>virsh vcpupin name-of-vm</code> command can be used. The output for a VM with 8 vCPU configured as above would look like this:
+To verify if the CPU threads have been successfully pinned to the VM the <code>virsh vcpupin name-of-vm</code> command can be used. The output for a VM with 8 vCPUs configured as above would look like this:
 <syntaxhighlight lang=text>
   VCPU   CPU Affinity

← Older revision		Revision as of 21:35, 1 March 2020
Line 119:		Line 119:

	=== Verifying CPU Affinity ===		=== Verifying CPU Affinity ===
		+	To verify if the CPU threads have been successfully pinned to the VM the <code>virst vcpupin name-of-vm</code> command can be used. The output for a VM wit 8 vCPU configured as above would look like this:
		+	<syntaxhighlight lang=text>
		+	VCPU CPU Affinity
		+	----------------------
		+	0 0
		+	1 1
		+	2 2
		+	3 3
		+	4 8
		+	5 9
		+	6 10
		+	7 11
		+	</syntaxhighlight>