Benchmark fetch and insert:

[seelabs]

These benchmarks time two operations:

1. The time to insert N values into a data base. The inserted keys and values are pseudo-randomly generated. The random number generator is always seeded with the same value for each run, so the same values are always inserted.
2. The time to fetch M existing values from a data base with N values. The order that the keys are fetched are pseudo-randomly generated. The random number generator is always seeded with the same value on each fun, so the keys are always looked up in the same order.

[seelabs]

@vinniefalco While there are more things I'd like to do with the benchmark (scritpt to generate graphs, benchmark dbs other than rocks, add more timings). I think this is good enough to commit and build off of.

[coveralls]

Coverage remained the same at 95.473% when pulling 99535b7e415cbfc469feaccdff7571085a221f22 on seelabs:bench into 8e2d542af438ee9b1a8d66bc78e1da41754fca38 on vinniefalco:develop.

[codecov-io]

Current coverage is 95.53% (diff: 100%)

Merging #27 into develop will not change coverage

@@            develop        #27   diff @@
==========================================
  Files            27         27          
  Lines          1746       1746          
  Methods           0          0          
  Messages          0          0          
  Branches          0          0          
==========================================
  Hits           1668       1668          
  Misses           78         78          
  Partials          0          0          

Powered by Codecov. Last update 0a6a28a...47a45b6

[vinniefalco]

Wow @ Those Numbers!

[coveralls]

Coverage remained the same at 95.533% when pulling 47a45b69ecc3c098be4da4e5979a98989ef33c22 on seelabs:bench into 0a6a28a6b6af79b84df4e89236dc5d8ce81d3a97 on vinniefalco:develop.

[vinniefalco]

Part of #28 now