Log in

No account? Create an account

Release 0.27 of Libmemcached

« previous entry | next entry »
Mar. 30th, 2009 | 08:21 am

So we have completed a major new version of libmemcached:

Added new UDP fire-forget mode.

This is in line with the model that Facebook uses. We have added a fire and forget mode for operations which can just be "sent" to the server. Depending on your client application and the size of your data you may be greatly surprised as to how much performance you can get out of this.

Reworked performance for mget() to better make use of async protocol

We have known for a while that our mget() operation while being fast, could be made faster. Instead of now cycling through servers we poll on data coming back from the server. This allows us to skip more "wait" cycles while data is being sent.

Cleaned up execution of fetch (just one set of code now)

This moves us to a cleaner code base. Some of the same thinking we have been doing in Drizzle. Create one execution path and build all operations on top of it. This gives us a smaller code base and removes the likelihood of bugs.

Fixed Jenkin's for big endian hosts.

Pretty simple here, the Jenkin's hash algorithm is now supported on big endian machines.

Updates for memstat to determine network latency.

Toru Maesaka has extended memstat to look for network latency (this was for Mixi.jp)

Updates for binary protocol.

Memcached recently released a new beta which supports the new binary protocol. (http://memcached.googlecode.com/files/memcached-1.3.2.tar.gz). The new protocol has shown to have some slight performance enhancements and allows for a host of new features.

You can find out more here:

Link | Leave a comment |

Comments {0}