We've set up a i3.4xlarge instance to migrate to a later version of Windows server for upgrade reasons. We've been running tests on the machine, before we cut it over to production, and the instance store SSD's were working perfectly last week in us-east1a but then something happened. This week, jobs that were running in under 5 seconds are now running 50+ seconds.
Big jobs that took an hour or so to crunch through are now running for 7 or 8 hrs.
We stopped the instance several times, thinking it might be underlying hardware issues but the performance has not come back.
The drives were already attached to the instance, as per usual. We just initialized and formatted them in the usual way.
Are there any tips or tricks to troubleshoot why the performance dropped and how to fix it?
I've been using AWS since 2008, so this isn't my first rodeo but, I've never seen this type of behavior.
Thanks,
Mike
AWS 官方已更新 4 個月前