Ceph

04/04/2018

02:12 PM Ceph Revision c7ade3a2 (ceph): kv/RocksDBStore: convert get() to use rocksdb Get() instead of iterator

For RocksDB, iterator will bypass row cache.
We don't want this to happen for get(), after row cache is enabled.
Sig... Jianjian Huo

10/05/2017

03:01 AM Ceph Revision 13360284 (ceph): kv/RocksDBStore: add CF support

- detect and use existing CFs on open
Signed-off-by: Jianjian Huo <jianjian.huo@ssi.samsung.com>
Signed-off-by: Adam... Jianjian Huo

10/03/2017

06:46 PM Ceph Revision 6bddacbf (ceph): kv/KeyValueDB: add column family interface

Signed-off-by: Jianjian Huo <jianjian.huo@ssi.samsung.com>
Signed-off-by: Sage Weil <sage@redhat.com> Jianjian Huo

09/28/2017

09:32 PM Ceph Revision fbf30e90 (ceph): kv/RocksDBStore: convert get() to use rocksdb Get() instead of iterator

For RocksDB, iterator will bypass row cache.
We don't want this to happen for get(), after row cache is enabled.
Sig... Jianjian Huo

06/17/2016

03:22 PM Ceph Revision f6ac5579 (ceph): kv/RocksDBStore: convert get() to use rocksdb Get() instead of iterator

For RocksDB, iterator will bypass row cache.
We don't want this to happen for get(), after row cache is enabled.
Sig... Jianjian Huo

06/10/2016

09:33 PM Ceph Revision 91db68c1 (ceph): os/bluestore: Add an option to choose bluefs direct or buffered io

For now, set default to buffered io; after rocksdb block cache
has ability to cache writes, we will set default to di... Jianjian Huo

06/07/2016

01:11 AM Ceph Revision 5bb63ebd (ceph): os/bluestore/BlueFS: convert all read/write IO to direct IO.

RocksDB is the sole user of BlueFS, after BlueFS uses direct IO,
RocksDB will have to use its own block cache to cach... Jianjian Huo

01:11 AM Ceph Revision 8e589130 (ceph): os/bluestore: convert read_buffered() to read_random() with buffered flag

New function will handle unaligned random read for both buffered or
direct read.
Signed-off-by: Jianjian Huo <jianji... Jianjian Huo

06/06/2016

09:44 PM Ceph Revision 17129122 (ceph): RocksDBStore: add option to configure number of block cache shards

rocksdb_cache_size will be splitted into number of shards evenly.
More shards will reduce lock contentions.
Signed-o... Jianjian Huo

04/27/2016

11:24 PM Ceph Revision 7e0953b5 (ceph): .mailmap and .orgmap: organization affiliation

Signed-off-by: Jianjian Huo <jianjian.huo@ssi.samsung.com> Jianjian Huo