Slurm's jobs in llmgw's accounting: projects as accounts, jobs priced
CI / test (push) Failing after 7m24s
CI / test (push) Failing after 7m24s
With slurm.users llmgw, each project is a Slurm account, its organization's and team's above it, with the project's members and its team's and organization's on it; slurm.conf holds a job to its user's accounts. The controller's agent keeps them with sacctmgr, only what it made, and reports the jobs that ended in the last day; the leader records each once, its GPU-hours (core-hours without cards) priced at auth.gpu_hour_cost into its project's usage as model slurm/PARTITION. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
1 parent
8547528fb7
commit
fcda4d3f85
22 files changed
+1390
-76
No files matched your search
@@ -0,0 +1,20 @@
|
||||
-- Ported from the SQLite migration of the same number.
|
||||
-- The Slurm jobs that ended, each recorded once (job_key is the cluster's
|
||||
-- name and the job's id), with what it used and cost; a usage row (model
|
||||
-- "slurm") carries the cost into the usage pages with the requests'.
|
||||
CREATE TABLE slurm_jobs (
|
||||
job_key TEXT PRIMARY KEY,
|
||||
user_id TEXT NOT NULL DEFAULT '',
|
||||
project_id TEXT NOT NULL DEFAULT '',
|
||||
login TEXT NOT NULL DEFAULT '',
|
||||
account TEXT NOT NULL DEFAULT '',
|
||||
partition TEXT NOT NULL DEFAULT '',
|
||||
state TEXT NOT NULL DEFAULT '',
|
||||
started_at BIGINT NOT NULL DEFAULT 0,
|
||||
ended_at BIGINT NOT NULL DEFAULT 0,
|
||||
elapsed_seconds BIGINT NOT NULL DEFAULT 0,
|
||||
cpus BIGINT NOT NULL DEFAULT 0,
|
||||
gpus BIGINT NOT NULL DEFAULT 0,
|
||||
cost_micros BIGINT NOT NULL DEFAULT 0
|
||||
);
|
||||
CREATE INDEX slurm_jobs_by_project ON slurm_jobs(project_id, ended_at);
|
||||
Reference in new issue
Block a user