daos-stack
DAOS-11318 bio: one bulk handle per DMA chunk
Mercury is able to support concurrent bulk transfers against single bulk handle, so the server bulk cache could be improved to create single bulk handle per DMA chunk, and all bulk transfers over the same chunk will share the same handle.
With this change, fewer handles will be created and more importantly, bulk chunk depopulate/populate doesn't need to destroy/create handles anymore, that'll be helpful for dealing with the potential bulk cache thrashing problem caused by complex IO patterns.
Minor fix to BIO metrics to count only the non-empty bulk groups.
Required-githooks: true
Signed-off-by: Niu Yawei [email protected]
Bug-tracker data: Ticket title is 'Use less bulk handles for server bulk cache ' Status is 'Open' Job should run at elevated priority (3) https://daosio.atlassian.net/browse/DAOS-11318
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
Test stage Functional Hardware Medium completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-9963/1/execution/node/965/log
Test stage Functional Hardware Medium completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-9963/2/execution/node/1059/log