Gridgain Query Local Testing Observations

Results of querying with varying number of records and number of nodes. I am generating some "Person" data and populating that into the grid. I am then making some queries on the data - Number (Big decimal) range comparisons and String comparisons - to get data or get counts. I am seeing as the whatever queries I am making, the time taken increases with the number of gridgain nodes I work with. This is true with count queries as well. I am not able to explain the result

The results are available in this pdf document.

Details of my machine Processor - Intel® Core™ i5-2520M CPU @ 2.50GHz × 4 Memory - 7.7 GiB
OS - Ubuntu - Release 12.04 (precise) 64-bit