Backlog #3523
Storing real CPU utilization (accounting record)
| Status: | Pending | Start date: | 01/22/2015 | |
|---|---|---|---|---|
| Priority: | Normal | Due date: | ||
| Assignee: | - | % Done: | 0% | |
| Category: | Drivers - Monitor | |||
| Target version: | - |
Description
(On behalf of the EGI Federated Cloud)
Both users and infrastructure providers are interested in long-term stats on real CPU utilization, ideally as part of the existing accounting records. An aggregated value for the actually spent CPU time would be very useful and wouldn't require excessive amounts of storage space (keeping all monitoring information would). Is there an easy way to do this?
Thanks!
History
#1
Updated by Ruben S. Montero over 6 years ago
- Tracker changed from Feature to Backlog
- Category changed from Core & System to Drivers - Monitor
This can be added as part of the monitoring information and it would be added to the VM template. Also note that this value MUST be aggregated as it each monitor probe will overrride the previous and just the last one is stored (i.e. same behavior as NET_TX and NET_RX). Moving this to backlog, we'll be happy to include this kind of probe to the stock monitoring scripts. Keep us updated if you make any progress in that direction.