Performance Comparison #13
Description
Right now, the buffer pool implementation is very similar to the one described in this paper by the Germans. The thing is, however, that I made several design decisions that made this buffer pool different. Additionally, there are a bunch of things that are still not implemented.
It would be silly to try and compare performance of this buffer pool manager against the German's numbers, but it might be worthwhile to compare this implementation against other storage managers like WiredTiger and RocksDB. If we can show that this asynchronous bpm is much more capable of exploiting parallelism than these other two, then we could really show that what we are working on is a worthwhile contribution.
I'm not really sure how to begin performance testing, let alone correctness testing, so I'm opening this issue. Since this is a buffer pool manager, it will likely be the case that TPC-C is the benchmark to use instead of TPC-H.