Anyone using Amazon EC2 small instances for a production cluster under reasonably high loads (perhaps hundreds or low thousands ops/sec)?

I'm looking for an example storage-conf.xml file to use as a guide. If you've done one and would be willing to share, please let me know.

I'm willing to bet a few people think running a few small nodes in a cluster is a bad idea and that maybe I should just use one or two large instances, but it should be doable for the throughput I'm targeting. Using smalls also provides a better spread of data (more smaller nodes equals more cuts of the data, which means less data to move after node failure).

- Andrew