[SERVER-44928] Add the large scale model to system_perf.yml Created: 03/Dec/19 Updated: 03/Dec/19 Resolved: 03/Dec/19 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Improvement | Priority: | Major - P3 |
| Reporter: | Donald Anderson | Assignee: | Donald Anderson |
| Resolution: | Won't Do | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Sprint: | Storage Engines 2019-12-16 |
| Participants: |
| Description |
|
The large scale model is the basic workload for the Large Scale Workload Automation project. In its current form, it runs under genny for 70 minutes (after populating data sets) and genny will process metrics which can take a few hours. All told, anywhere from 2 to 7 hours (as genny can OOM during the process). Still, this is valuable for the storage team when WiredTiger metrics are captured. And as genny becomes more capable, we will extend the run time to many hours or days. So clearly, this needs to be its own workload job. Piggybacking onto genny_auto_tasks or genny_patch_tasks would be wasteful and confusing when selected. |
| Comments |
| Comment by Donald Anderson [ 03/Dec/19 ] |
|
ryan.timmons and I discussed this, and decided that we'll put this into the genny AutoRun mechanism. Even though it may run many hours, and may fail a lot, we hope to have enough successes to find what our baseline and variance is. See https://jira.mongodb.org/browse/TIG-2201 . |