On 13 feb, 12:31, Alan Williamson <[email protected]> wrote: > The key difference with DynamoDB and > SimpleDB is that for any large cross-document queries you wish to do, > you really have to jump into MapReduce algorithms to get any sort of > performance from it.
Alan, I understand that DynamoDB has big performance/scale/reliable advantages over SimpleDB, but for datasets that will never hit the 10Gb and need rather simple queries like "all users from organisation X", solutions that you would need to build with DynamoDB sound complex and a little over the top. Like building your own index in seperate tables. Doable, but not easy. I would very much like to "jump into MapReduce algorithms" (in fact, if you have a link to example implementation, I would be very interested), but I don't know that and that doesn't sound easy either, while the solution in SimpleDB is. Or am I missing something? Do you think SimpleDB could still work in specific scenarios or should be simply avoid it? I have read articles from others putting search indexes with e.g. SOLR in front of DynamoDB. Search via search engines, followed by a retrieval in DynamoDB. That sounds like a good approach. In fact, I have read suggestions that Amazon is already working on a search- service itself. And back to the original question .... any plans to support DynamoDB in openBD? -- online documentation: http://openbd.org/manual/ google+ hints/tips: https://plus.google.com/115990347459711259462 http://groups.google.com/group/openbd?hl=en Join us @ http://www.OpenCFsummit.org/ Dallas, Feb 2012
