Job Usage Calculation
The raw job usage factor for an association is defined as the sum of products
of number of nodes used (nnodes) and time elapsed (t_elapsed). To
calculate the raw usage for a given association U:
\(U = sum(nnodes \times t\_elapsed)\)
flux-accounting keeps track of job usage in a table according to two properties that are set when the database is first created: PriorityDecayHalfLife and PriorityUsageResetPeriod. Each of these parameters represent a number of weeks by which to hold usage factors up to the time period where jobs no longer play a factor in calculating a usage factor. If these options aren't specified, the table defaults to 4 usage columns, each which represent one week's worth of jobs.
The job usage factor table stores past job usage factors per association. When an association is first added to the association table, they are also added to to job usage factor table.
The value of PriorityDecayHalfLife determines the amount of time that
represents one "usage period" of jobs. flux-accounting filters out its jobs
table and retrieves an association's jobs that have completed in the usage
period.
As time goes on and usage periods get older, their raw usage value has a decay factor \(D\) (0.5) applied to them before they are added to the user's current raw usage factor.
\(U_{past} = (D \times U_{last\_period}) + (D \times D \times U_{period-2}) + ...\)
After the current usage factor is calculated, it is written to the first usage bin in the job usage factor table along with the other, older factors. The oldest factor then gets removed from the table since it is no longer needed.
An example
Let's say an association has the following job records from the most recent PriorityDecayHalfLife:
UserID Username JobID T_Submit T_Run T_Inactive Nodes R
0 1002 1002 102 1605633403.22141 1605635403.22141 1605637403.22141 2 {"version":1,"execution": {"R_lite":[{"rank":"0","children": {"core": "0"}}]}}
1 1002 1002 103 1605633403.22206 1605635403.22206 1605637403.22206 2 {"version":1,"execution": {"R_lite":[{"rank":"0","children": {"core": "0"}}]}}
2 1002 1002 104 1605633403.22285 1605635403.22286 1605637403.22286 2 {"version":1,"execution": {"R_lite":[{"rank":"0","children": {"core": "0"}}]}}
3 1002 1002 105 1605633403.22347 1605635403.22348 1605637403.22348 1 {"version":1,"execution": {"R_lite":[{"rank":"0","children": {"core": "0"}}]}}
4 1002 1002 106 1605633403.22416 1605635403.22416 1605637403.22416 1 {"version":1,"execution": {"R_lite":[{"rank":"0","children": {"core": "0"}}]}}
total nodes used: 8
total time elapsed: 10000.0
\(U_{user1002\_current}\) is calculated as:
\(U_{user1002\_current} = (2 \times 2000) + (2 \times 2000) + (2 \times 2000) + (1 \times 2000) + (1 \times 2000)\)
\(U_{user1002\_current} = 4000 + 4000 + 4000 + 2000 + 2000\)
\(U_{user1002\_current} = 16000\)
And the association's past job usage factors (each one represents a PriorityDecayHalfLife period up to the PriorityUsageResetPeriod) consists of the following:
username bank usage_factor_period_0 usage_factor_period_1 usage_factor_period_2 usage_factor_period_3
0 user1002 C 128.0000 64.00000 64.0000 16.00000
The past usage factors have the decay factor applied to them:
[64.0, 16.0, 8.0, 1.0]
\(U_{user1002\_past} = 64.0 + 16.0 + 8.0 + 1.0 = 89\)
\(U_{user1002\_historical} = U_{user1002\_current} + U_{user1002\_past} = 16000.0 + 89.0 = 16089.0\)
\(U_{user1002}\)'s job usage value now becomes \(16089.0\), which takes into account both their most recent and historical job usage.