Page MenuHomePhabricator

Figure out goals of six week test for labs replica of elasticsearch index
Closed, DuplicatePublic


mark approved a six week test with a machine that will be available in labs and recieve writes from prod.

Server specs: Dell PowerEdge R420, dual Intel Xeon E5-2450 v2 2.50GHz, 64GB Memory, (4) 3TB Disks

Event Timeline

EBernhardson claimed this task.
EBernhardson raised the priority of this task from to Needs Triage.
EBernhardson updated the task description. (Show Details)
Restricted Application added a subscriber: Aklapper. · View Herald TranscriptSep 18 2015, 5:34 PM

We need to figure out how exactly we want the disks configured. The prod indices, in total, are currently 2.4TB. We could cut that down to 1.2TB by only taking content indices if necessary (although the support for sending only some indices to another cluster does not yet exist, but could be created). The prod cluster currently handles writes of between 2 and 10 documents per second, this machine will need to do the same. I wouldn't worry too much about this though in the previous labs test ingestion didn't show any problems.

In terms of query performance we know we will be reading from disk. A lot. If our dataset was smaller i would almost prefer to set this disks up in a raid 1. To fit everything in maybe software raid 10 would give us what we need?