Bug #11322: rgw - improve performance for large object (multiple chunks) GET - rgw - Ceph

Bug #11322

rgw - improve performance for large object (multiple chunks) GET

Added by Guang Yang over 8 years ago. Updated almost 6 years ago.

Status:

Resolved

Priority:

Normal

Assignee:

Guang Yang

Target version:

% Done:

Source:

other

Tags:

Backport:

hammer

Regression:

Severity:

3 - minor

Reviewed:

Affected Versions:

ceph-qa-suite:

Pull request ID:

Crash signature (v1):

Crash signature (v2):

Description

Currently the handling at rgw side for large object (multiple chunks) is, it sends response (if there is any) first and then issues AIO to OSD for next object, one optimization is to switch that, to make get from OSD and response to client parallel.