And a related question... Would there be an advantage in using lazy
byte strings or Bulat's streams library over HDF5? There's a good
performance review  of PyTables (thin wrapper on top of HDF5) vs.
SqlLite that I just read and it made me wonder.
I'm looking to model the trading process as (lazy?) stream processing
where a trade represents an event in the input stream and trade and
stop orders would be injected into the output stream.