A series of microbenchmarks for jsonschema in Pythons (2.7.5, 3.3.2, pypy-2.1).

jsonschema-benchmarks.py

# Microbenchmarks using jsonschema in Python (2.7.5, 3.3.2) and pypy 2.1

## Common code
import jsonschema 

schema = {
  'type': 'object',
  'properties': {
    'weight': {'type': 'integer', 'minimum': 1},
    'hosts': {'type': 'array', 'minItems': 1, 'items':
      {'type': 'string'}
  }
}

## pypy 2.1

### JIT warm up run
In [6]: %timeit jsonschema.validate({'weight': 1}, schema)
100 loops, best of 3: 1.63 ms per loop

In [7]: %timeit jsonschema.validate({'weight': 1}, schema)
1000 loops, best of 3: 259 µs per loop

In [10]: validator = jsonschema.Draft4Validator(schema)

In [11]: %timeit validator.validate({'weight': 1})
10000 loops, best of 3: 7.88 µs per loop

In [12]: %timeit validator.validate({'weight': 1})
10000 loops, best of 3: 7.93 µs per loop


## Python 3.3.2
In [3]: %timeit jsonschema.validate({'weight': 1}, schema)
1000 loops, best of 3: 1.44 ms per loop

In [4]: %timeit jsonschema.validate({'weight': 1}, schema)
1000 loops, best of 3: 1.44 ms per loop

In [5]: validator = jsonschema.Draft4Validator(schema)

In [6]: %timeit validator.validate({'weight': 1})
10000 loops, best of 3: 32.9 us per loop

In [7]: %timeit validator.validate({'weight': 1})
10000 loops, best of 3: 33.1 us per loop

## Python 2.7.5

In [3]: %timeit jsonschema.validate({'weight': 1}, schema)
1000 loops, best of 3: 1.14 ms per loop

In [4]: validator = jsonschema.Draft4Validator(schema)

In [5]: %timeit validator.validate({'weight': 1})
10000 loops, best of 3: 35.3 µs per loop