From: Daniel Kral <d.kral@proxmox.com>
To: pve-devel@lists.proxmox.com
Subject: [PATCH proxmox v3 08/40] resource-scheduling: factor out topsis alternative mapping
Date: Mon, 30 Mar 2026 16:30:17 +0200 [thread overview]
Message-ID: <20260330144101.668747-9-d.kral@proxmox.com> (raw)
In-Reply-To: <20260330144101.668747-1-d.kral@proxmox.com>
The same calculation will be needed for the scoring of migrations with
the TOPSIS method in the following patch.
Signed-off-by: Daniel Kral <d.kral@proxmox.com>
Reviewed-by: Dominik Rusovac <d.rusovac@proxmox.com>
---
changes v2 -> v3:
- none
proxmox-resource-scheduling/src/scheduler.rs | 68 ++++++++++++--------
1 file changed, 42 insertions(+), 26 deletions(-)
diff --git a/proxmox-resource-scheduling/src/scheduler.rs b/proxmox-resource-scheduling/src/scheduler.rs
index 33ef4586..5aca549d 100644
--- a/proxmox-resource-scheduling/src/scheduler.rs
+++ b/proxmox-resource-scheduling/src/scheduler.rs
@@ -44,6 +44,44 @@ impl Scheduler {
}
}
+ /// Map the current node usages to a [`PveTopsisAlternative`].
+ ///
+ /// The [`PveTopsisAlternative`] is derived by calculating a modified version of the root mean
+ /// square (RMS) and maximum value of each stat in the node usages.
+ fn topsis_alternative_with(
+ &self,
+ map_node_stats: impl Fn(&NodeUsage) -> NodeStats,
+ ) -> PveTopsisAlternative {
+ let len = self.nodes.len();
+
+ // Base values on percentages to allow comparing nodes with different stats.
+ let mut highest_cpu = 0.0;
+ let mut squares_cpu = 0.0;
+ let mut highest_mem = 0.0;
+ let mut squares_mem = 0.0;
+
+ for node in self.nodes.iter() {
+ let new_stats = map_node_stats(node);
+
+ let new_cpu = new_stats.cpu_load();
+ highest_cpu = f64::max(highest_cpu, new_cpu);
+ squares_cpu += new_cpu.powi(2);
+
+ let new_mem = new_stats.mem_load();
+ highest_mem = f64::max(highest_mem, new_mem);
+ squares_mem += new_mem.powi(2);
+ }
+
+ // Add 1.0 to avoid boosting tiny differences: e.g. 0.004 is twice as much as 0.002, but
+ // 1.004 is only slightly more than 1.002.
+ PveTopsisAlternative {
+ average_cpu: 1.0 + (squares_cpu / len as f64).sqrt(),
+ highest_cpu: 1.0 + highest_cpu,
+ average_memory: 1.0 + (squares_mem / len as f64).sqrt(),
+ highest_memory: 1.0 + highest_mem,
+ }
+ }
+
/// Scores nodes to start a resource with the usage statistics `resource_stats` on.
///
/// The scoring is done as if the resource is already started on each node. This assumes that
@@ -56,43 +94,21 @@ impl Scheduler {
&self,
resource_stats: T,
) -> Result<Vec<(String, f64)>, Error> {
- let len = self.nodes.len();
let resource_stats = resource_stats.into();
let matrix = self
.nodes
.iter()
.map(|node| {
- // Base values on percentages to allow comparing nodes with different stats.
- let mut highest_cpu = 0.0;
- let mut squares_cpu = 0.0;
- let mut highest_mem = 0.0;
- let mut squares_mem = 0.0;
-
- for target_node in self.nodes.iter() {
+ self.topsis_alternative_with(|target_node| {
let mut new_stats = target_node.stats;
if node.name == target_node.name {
new_stats.add_started_resource(&resource_stats)
- };
+ }
- let new_cpu = new_stats.cpu_load();
- highest_cpu = f64::max(highest_cpu, new_cpu);
- squares_cpu += new_cpu.powi(2);
-
- let new_mem = new_stats.mem_load();
- highest_mem = f64::max(highest_mem, new_mem);
- squares_mem += new_mem.powi(2);
- }
-
- // Add 1.0 to avoid boosting tiny differences: e.g. 0.004 is twice as much as 0.002, but
- // 1.004 is only slightly more than 1.002.
- PveTopsisAlternative {
- average_cpu: 1.0 + (squares_cpu / len as f64).sqrt(),
- highest_cpu: 1.0 + highest_cpu,
- average_memory: 1.0 + (squares_mem / len as f64).sqrt(),
- highest_memory: 1.0 + highest_mem,
- }
+ new_stats
+ })
.into()
})
.collect::<Vec<_>>();
--
2.47.3
next prev parent reply other threads:[~2026-03-30 14:43 UTC|newest]
Thread overview: 72+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-30 14:30 [PATCH-SERIES cluster/ha-manager/perl-rs/proxmox v3 00/40] dynamic scheduler + load rebalancer Daniel Kral
2026-03-30 14:30 ` [PATCH proxmox v3 01/40] resource-scheduling: inline add_cpu_usage in score_nodes_to_start_service Daniel Kral
2026-03-31 6:01 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH proxmox v3 02/40] resource-scheduling: move score_nodes_to_start_service to scheduler crate Daniel Kral
2026-03-31 6:01 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH proxmox v3 03/40] resource-scheduling: rename service to resource where appropriate Daniel Kral
2026-03-31 6:02 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH proxmox v3 04/40] resource-scheduling: introduce generic scheduler implementation Daniel Kral
2026-03-31 6:11 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH proxmox v3 05/40] resource-scheduling: implement generic cluster usage implementation Daniel Kral
2026-03-31 7:26 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH proxmox v3 06/40] resource-scheduling: topsis: handle empty criteria without panics Daniel Kral
2026-03-30 14:30 ` [PATCH proxmox v3 07/40] resource-scheduling: compare by nodename in score_nodes_to_start_resource Daniel Kral
2026-03-30 14:30 ` Daniel Kral [this message]
2026-03-30 14:30 ` [PATCH proxmox v3 09/40] resource-scheduling: implement rebalancing migration selection Daniel Kral
2026-03-31 7:33 ` Dominik Rusovac
2026-03-31 12:42 ` Michael Köppl
2026-03-31 13:32 ` Daniel Kral
2026-03-30 14:30 ` [PATCH perl-rs v3 10/40] pve-rs: resource-scheduling: remove pedantic error handling from remove_node Daniel Kral
2026-03-30 14:30 ` [PATCH perl-rs v3 11/40] pve-rs: resource-scheduling: remove pedantic error handling from remove_service_usage Daniel Kral
2026-03-30 14:30 ` [PATCH perl-rs v3 12/40] pve-rs: resource-scheduling: move pve_static into resource_scheduling module Daniel Kral
2026-03-30 14:30 ` [PATCH perl-rs v3 13/40] pve-rs: resource-scheduling: use generic usage implementation Daniel Kral
2026-03-31 7:40 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH perl-rs v3 14/40] pve-rs: resource-scheduling: static: replace deprecated usage structs Daniel Kral
2026-03-30 14:30 ` [PATCH perl-rs v3 15/40] pve-rs: resource-scheduling: implement pve_dynamic bindings Daniel Kral
2026-03-30 14:30 ` [PATCH perl-rs v3 16/40] pve-rs: resource-scheduling: expose auto rebalancing methods Daniel Kral
2026-03-30 14:30 ` [PATCH cluster v3 17/40] datacenter config: restructure verbose description for the ha crs option Daniel Kral
2026-03-30 14:30 ` [PATCH cluster v3 18/40] datacenter config: add dynamic load scheduler option Daniel Kral
2026-03-30 14:30 ` [PATCH cluster v3 19/40] datacenter config: add auto rebalancing options Daniel Kral
2026-03-31 7:52 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH ha-manager v3 20/40] env: pve2: implement dynamic node and service stats Daniel Kral
2026-03-31 13:25 ` Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 21/40] sim: hardware: pass correct types for static stats Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 22/40] sim: hardware: factor out static stats' default values Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 23/40] sim: hardware: fix static stats guard Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 24/40] sim: hardware: handle dynamic service stats Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 25/40] sim: hardware: add set-dynamic-stats command Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 26/40] sim: hardware: add getters for dynamic {node,service} stats Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 27/40] usage: pass service data to add_service_usage Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 28/40] usage: pass service data to get_used_service_nodes Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 29/40] add running flag to non-HA cluster service stats Daniel Kral
2026-03-31 7:58 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH ha-manager v3 30/40] usage: use add_service to add service usage to nodes Daniel Kral
2026-03-31 8:12 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH ha-manager v3 31/40] usage: add dynamic usage scheduler Daniel Kral
2026-03-31 8:15 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH ha-manager v3 32/40] test: add dynamic usage scheduler test cases Daniel Kral
2026-03-31 8:20 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH ha-manager v3 33/40] manager: rename execute_migration to queue_resource_motion Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 34/40] manager: update_crs_scheduler_mode: factor out crs config Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 35/40] implement automatic rebalancing Daniel Kral
2026-03-31 9:07 ` Dominik Rusovac
2026-03-31 9:07 ` Michael Köppl
2026-03-31 9:16 ` Dominik Rusovac
2026-03-31 9:32 ` Daniel Kral
2026-03-31 9:39 ` Dominik Rusovac
2026-03-31 13:55 ` Daniel Kral
2026-03-31 9:42 ` Daniel Kral
2026-03-31 11:01 ` Michael Köppl
2026-03-31 13:50 ` Daniel Kral
2026-03-30 14:30 ` [PATCH ha-manager v3 36/40] test: add resource bundle generation test cases Daniel Kral
2026-03-31 9:09 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH ha-manager v3 37/40] test: add dynamic automatic rebalancing system " Daniel Kral
2026-03-31 9:33 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH ha-manager v3 38/40] test: add static " Daniel Kral
2026-03-31 9:44 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH ha-manager v3 39/40] test: add automatic rebalancing system test cases with TOPSIS method Daniel Kral
2026-03-31 9:48 ` Dominik Rusovac
2026-03-30 14:30 ` [PATCH ha-manager v3 40/40] test: add automatic rebalancing system test cases with affinity rules Daniel Kral
2026-03-31 10:06 ` Dominik Rusovac
2026-03-31 20:44 ` partially-applied: [PATCH-SERIES cluster/ha-manager/perl-rs/proxmox v3 00/40] dynamic scheduler + load rebalancer Thomas Lamprecht
2026-04-02 12:55 ` superseded: " Daniel Kral
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260330144101.668747-9-d.kral@proxmox.com \
--to=d.kral@proxmox.com \
--cc=pve-devel@lists.proxmox.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox