How to save 200% RAM by selecting the right key data type for #MongoDB

gistfile1.bash

Intel(R) Core(TM) i3 CPU       M 350  @ 2.27GHz
Mem 2G
bvcom@ubuntu:~/work/db/test$ time python mongoIndexTest.py 
ObjectID
int
Base64 BSON
string

real	4m59.993s
user	2m30.729s
sys	2m14.420s


> show collections
base64s
ints
objectids
strings
system.indexes
> db.ints.stats()
{
	"ns" : "test.ints",
	"count" : 1000000,
	"size" : 24000040,
	"avgObjSize" : 24.00004,
	"storageSize" : 43057152,
	"numExtents" : 11,
	"nindexes" : 1,
	"lastExtentSize" : 10440704,
	"paddingFactor" : 1,
	"flags" : 1,
	"totalIndexSize" : 32522240,
	"indexSizes" : {
		"_id_" : 32522240
	},
	"ok" : 1
}
> db.objectids.stats()
{
	"ns" : "test.objectids",
	"count" : 1000000,
	"size" : 32000032,
	"avgObjSize" : 32.000032,
	"storageSize" : 55586816,
	"numExtents" : 12,
	"nindexes" : 1,
	"lastExtentSize" : 12529664,
	"paddingFactor" : 1,
	"flags" : 1,
	"totalIndexSize" : 41598976,
	"indexSizes" : {
		"_id_" : 41598976
	},
	"ok" : 1
}
> db.base64s.stats()
{
	"ns" : "test.base64s",
	"count" : 1000000,
	"size" : 40000032,
	"avgObjSize" : 40.000032,
	"storageSize" : 70623232,
	"numExtents" : 13,
	"nindexes" : 1,
	"lastExtentSize" : 15036416,
	"paddingFactor" : 1,
	"flags" : 1,
	"totalIndexSize" : 67117056,
	"indexSizes" : {
		"_id_" : 67117056
	},
	"ok" : 1
}
> db.strings.stats()
{
	"ns" : "test.strings",
	"count" : 1000000,
	"size" : 56000016,
	"avgObjSize" : 56.000016,
	"storageSize" : 88670208,
	"numExtents" : 14,
	"nindexes" : 1,
	"lastExtentSize" : 18046976,
	"paddingFactor" : 1,
	"flags" : 1,
	"totalIndexSize" : 90267648,
	"indexSizes" : {
		"_id_" : 90267648
	},
	"ok" : 1
}

mongoIndexTest.py

#!/usr/bin/env python
#coding: utf8
# This Code copy from AlexDong.com
# http://notes.alexdong.com/choose-the-right-data-type-for-mongodb

import pymongo
import bson
import hashlib
from pymongo import Connection

db = Connection().test

count = 1000000
print('ObjectID')
for i in range(0, count):
    db.objectids.insert({'i': i})

print('int')
for i in range(0, count):
    db.ints.insert({'_id': i, 'i': i})

print('Base64 BSON')
for i in range(0, count):
    db.base64s.insert({'_id': \
        bson.Binary(hashlib.md5(str(i)).digest(),
        bson.binary.MD5_SUBTYPE), 'i': i})

print('string')
for i in range(0, count):
    db.strings.insert({'_id': hashlib.md5(str(i)).hexdigest(), 'i': i})