[SERVER-45866] Measure connection reuse for Hedged Reads Created: 30/Jan/20  Updated: 16/Apr/20  Resolved: 16/Apr/20

Status: Closed
Project: Core Server
Component/s: Sharding
Affects Version/s: None
Fix Version/s: None

Type: Task Priority: Major - P3
Reporter: Misha Tyulenev Assignee: Benjamin Caimano (Inactive)
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: PNG File baton 30mins.png     PNG File baton.png     PNG File baton.png     PNG File primary.png     PNG File reuse.png     PNG File secondary.png    
Issue Links:
Related
related to SERVER-46512 Ensure that connections acquired for ... Closed
is related to SERVER-47602 _killOperations for hedged reads can ... Closed
Sprint: Service Arch 2020-02-10, Service Arch 2020-02-24, Service Arch 2020-03-09, Service Arch 2020-03-23, Service Arch 2020-04-06, Service arch 2020-04-20
Participants:

 Description   

Validate that a workload that consists of mostly maxtimems second queries achieves effective connection reuse. The notion here is that connection reuse after cancellation should prevent each operation from turning into a new connection establishment.

Perform performance tests against a 3 node replica set with nearest on

  • Stream in a steady amount of .5, 1, 2 and 5 second reads (without maxtimems) with read pref nearest, at no more than 10% of overall read capacity for any particular node
  • Track how often we are able to kill, then reuse connections

Generated at Thu Feb 08 05:09:54 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.