More notes on Membase and memcached
As a companion to my post about Membase last week, the company has graciously allowed me to post a rather detailed Membase slide deck. (It even has pricing.) Also, I left one point out.
Membase announced a Cloudera partnership. I couldn’t detect anything technically exciting about that, but it serves to highlight what I do find to be an interesting usage trend. A couple of big Web players (AOL and ShareThis) are using Hadoop to crunch data and derive customer profile data, then feed that back into Membase. Why Membase? Because it can serve up the profile in a millisecond, as part of a bigger 40-millisecond-latency request.
And why Hadoop, rather than Aster Data nCluster, which ShareThis also uses? Umm, I didn’t ask.
When I mentioned this to Colin Mahony, he said Vertica had similar stories. However, I don’t recall whether they were about Membase or just memcached, and he hasn’t had a chance to get back to me with clarification. (Edit: As per Colin’s comment below, it’s both.)
Comments
7 Responses to “More notes on Membase and memcached”
Leave a Reply
Lots of database servers have 1ms response times for cached data
Good to know. Loading data from Membase to Hadoop and vice verse effectively, or running M/R jobs directly from Membase – not so bad idea.
Mark,
Fair enough.
CAM
Curt,
There are several Vertica customers using Membase in the manner you describe with bi-directional integration back to Vertica- sharing scoring models for instance. I also know of several customers using Memcache for object caching and other uses.
cpm
Curt,
Here is an impressive result demonstrating that MySQL can be web-scale — 750,000 QPS from MySQL/InnoDB using a new interface. Hopefully this code will make it into a supported distribution.
http://yoshinorimatsunobu.blogspot.com/2010/10/using-mysql-as-nosql-story-for.html
Mark,
As I’ve noted elsewhere, I suspect that key-value has a performance advantage over relational, but that the advantage is no more than 2X.
But it’s pretty impossible to do a fair test, because nobody’s really motivated to put the same level of effort into optimizing each and then compare them side by side.
Besides, it all depends on specifics of the database and of the use case.
More notes on Membase and memcached …
As a companion to my post about Membase last week, the company has graciously allowed me to post a rather detailed Membase slide deck. (It even has pricing.) Also, I left one point out.
…