andy_pavlo

I really wanted to like etcd, but Andy Pavlo was right

Andy Pavlo of the CMU Database Group is well known for saying that while NoSQL databases acquire cyclical popularity, all databases eventually iterate back to a SQL interface — it happened with MongoDB and Google’s BigTable for example.

I think I have hit that point with etcd. Initially I ported from MySQL to etcd because I really wanted the inexpensive distributed locking and being able to watch values. However, I never actually watch values in my code any more, and I now spend a huge amount of my time maintaining what my code calls “caches”, but which I can now see are just poorly implemented secondary indexes. The straw that broke the camel’s back was https://github.com/etcd-io/etcd/issues/9043, which changed etcd’s defaults to only being able to return 1.5mb in a RPC request.

I therefore think it might be time for me to port back to a real SQL database, perhaps keeping etcd to manage distributed locks. Perhaps.

(more…)

Continue ReadingI really wanted to like etcd, but Andy Pavlo was right

Databases built on object stores are officially interesting

  • Post author:
  • Post category:Databases

…even if none of my friends seem to think so.

I’ve been off on a bit of a tangent recently. Its a slow burn tangent, that I am pretty sure was kicked off by this Geek Narrator podcast episode about the design of Turbo Puffer with Simon Eskildsen:

The basic idea is that you can build very large scale database systems using only the primitives provided by an object store such as Amazon S3. Now, the performance might also suck, but you can alleviate some of that with a good caching layer and in return you get massive scale. This first video caused me to discover the work of Andy Pavlo, who was interviewed by the same podcast:

(more…)

Continue ReadingDatabases built on object stores are officially interesting

End of content

No more pages to load