[pytables-users] Aggregate on criteria

[Ali hallaji]

Hi,
I want to fast aggregate on my result from "where", and I don't know how to
set iterator by 'numpy' or 'expr' on my result for fast computing.
Are you any solution for this question?
I want to use from PyTables for fast and realtime query.Are you advise me
about PyTables, that I use PyTables or MongoDB?
Are
The benchmark is in this context? PyTables vs MongoDB?
I set benchmark on write, which PyTables much faster than MongoDB(about 5x)
PyTables use filter('Blosc', level='5')

Write for PyTables on 10 Million complex rows = 21 min
Write for MongoDB on 10 Million complex rows = 118 min

Index on 'full_name' field on 10 Million case = 48 sec # PyTables {filters:
(Blosc, 5), kind: 'full', optlevel: 9}
Index on 'full_name' field on 10 Million case = 12 sec # PyTables {filters:
(Blosc, 5), kind: 'ultralight', optlevel: 9}

Index on 'full_name' field on 10 Million case = 2 min # MonogDB

I want to need other benchmark for aggregate.
Please guide me for better solution.