[LU-7990] Large bulk IO support

• Type: New Feature
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.9.0, Lustre 2.11.0, Lustre 2.10.2
• Labels: patch
• Assignee: Jinshan Xiong (Inactive)
• Reporter: Gu Zheng (Inactive)

Description

Add large bulk IO support, e.g. make the ptlrpc be able to size 16MB IO, to improve the performance.

Issue Links

• is related to: LU-4533 rpc_stats histogram does not support max_rpcs_in_flight greater than 31
• is related to: LU-8193 request mbits isn't set properly for EINPROGRESS resend
• is related to: LU-2957 Proc tnable to control RAS_INCREASE_STEP
• is related to: LU-7140 Few 4 MB BRW rpcs from readahead path
• is related to: LUDOC-356 Lustre Manual - Large Bulk IO
• is related to: LU-9356 add blk_plug into osd-ldiskfs when do bio

Comments

[Gerrit Updater]

John L. Hammond ([e-mail]) merged in patch https://review.whamcloud.com/29738/
Subject: LU-7990 llite: increase whole-file readahead to RPC size
Project: fs/lustre-release
Branch: b2_10
Current Patch Set:
Commit: 5c1877c4143bddcd3c5c4f2397db5622c2116fce

[Gerrit Updater]

Minh Diep ([e-mail]) uploaded a new patch: https://review.whamcloud.com/29738
Subject: LU-7990 llite: increase whole-file readahead to RPC size
Project: fs/lustre-release
Branch: b2_10
Current Patch Set: 1
Commit: 9482cd464b354cd1a3edddb158d42eea2d733419

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch https://review.whamcloud.com/26955/
Subject: LU-7990 llite: increase whole-file readahead to RPC size
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 627d0133d9d7bef908544537a8a2b2093719fc72

[Gerrit Updater]

Andreas Dilger ([e-mail]) uploaded a new patch: https://review.whamcloud.com/26955
Subject: LU-7990 llite: increase whole-file readahead to RPC size
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 25f80fc94b7988e9d356de200573027bc5178a32

[Evan D. Chen (Inactive)]

Add LUDOC-356 into lustre manual tracking for Large bulk IO feature.

[Joseph Gmitter (Inactive)]

Patches have landed to master for 2.9.0

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/19368/
Subject: LU-7990 clio: revise readahead to support 16MB IO
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: d8467ab8a2ca15fbbd5be3429c9cf9ceb0fa78b8

[James A Simmons]

Over time these larger allocations are going to fail due to the increasing memory fragmentation so the vmalloc penalty will show up.

[Alexey Lyashkov]

Andreas,

64k allocations failed easly after some time. If you look into several LNet OOM tickets you may see it. So any aged system including servers will switch to use vmalloc for this code.

As about SG lists - server side R/O cache kills an IB OFED cache as pool code can't find older map. I worked on other possibilities time to time. when looked to rework an o2ib LND.

[Jinshan Xiong (Inactive)]

Andreas - I'm reluctant to increase the initial grant because I would like to be conservative before 16M RPC is widely used. Without this change, the first RPC would be small in size, but after that the RPC size should be 16MB, not a big deal.