Hi, >I'm a bit skeptical if the added complexity is worth the disk space >saving.
As for disk space: when storing data as JSON (as I guess 99% of all users of MongoDB do), MongoDB compresses the data as well. Currently the format seems to be BSON, and there are plans to Snappy / LZO / LZ4. See also https://jira.mongodb.org/browse/SERVER-164 As for in-memory: it depends on what level we cache, on whether we cache Java objects (node objects / property objects) or a lower level representation. Caching a segment might not be efficient if only a small subset of the segment is accessed. Caching on a finer grained level (just nodes) might be more efficient. So I guess disk space and memory usage is just "different", but it's not so easy to say whether it's "better" or "worse". Regards, Thomas
