- Fixed one issue with native memory allocator used internally by Koda which greatly affected multithreaded update performance. The preliminary test numbers are ~ 4M queries per sec with 90/10 read/write ratio (which is up 15%). I think for 50/50 ratio performance gain must be much larger.
- Implemented persistence layer support (which is based on a very promising leveldb library). The support is not integrated yet (will be working on this later on next week)
- Currently working on compression support (snappy and gzip) for Koda.
Thursday, August 18, 2011
Update on Koda
Subscribe to:
Post Comments (Atom)
Hi Vladimir,
ReplyDeleteIs the source code of Koda going to be available in near future? Would love to see the design and use it for my personal project.
Thanks
It will be released to the public as a part of a new distributed k-v data store some time later on this year.
ReplyDeleteAny updates on releasing it for testing?
ReplyDelete12-31-11.
ReplyDeleteCool.
ReplyDeleteCant wait to get my hand on this thing.
Cool, in 2 day it's 31.12. I'm curious to test this tool too.
ReplyDeleteCould you pls comment, how is your solution compared to this one http://www.infoq.com/presentations/Concurrent-Caching-at-Google from Google.
Thank you!
Was this ever released? I am very interested in taking a look if so.
ReplyDeleteStill interested in seeing what comes out of this... how's progress going?
ReplyDelete