https://github.com/baseprime/mongobench

MongoDB Benchmarking
https://github.com/baseprime/mongobench

Last synced: 11 days ago
JSON representation

MongoDB Benchmarking

• Host: GitHub
• URL: https://github.com/baseprime/mongobench
• Owner: baseprime
• License: gpl-3.0
• Created: 2015-06-18T05:32:09.000Z (over 9 years ago)
• Default Branch: master
• Last Pushed: 2015-06-21T06:45:33.000Z (over 9 years ago)
• Last Synced: 2024-08-01T22:56:17.288Z (3 months ago)
• Language: Python
• Size: 348 KB
• Stars: 6
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
mongobench

===========

A benchmarking tool for MongoDB. The tool performs both read and write tests (using insert() and find()). Test data is generated by the tool based on the number of 'str_cnt' defined i.e. 1 ... n UUID string elements are created depending on the 'str_cnt' defined. The tool will measure the time taken for each read / write operation and provide statistics upon completion (including both average query time and queries per second).

The number of documents to be written and read can also be defined (in the 'documents' variable) as well as the number of concurrent connections that should be created by the process. Currently all options need to be defined within the script. 

Sample output:

```

WRITE TEST RESULTS (insert):

Number of documents: 1024

Concurrency: 64

Number of benchmarks: 1

Average connection & query time: 0.0249066145625

Average QPS:  2408.99861559

READ TEST RESULTS (find):

Number of documents: 1024

Concurrency: 64

Number of benchmarks: 1

Average connection & query time: 0.00104909064248

Average QPS:  57192.3888846

```

The variables that need to be configured are:

```

database = 'mongobench' # The database to be used for benchmarking

collection = 'testcollection' # The default collection prefix name

documents = 100000 # The number of documents to be written and read

benchmarks = 1 # The number of iterations for each type of benchmark (Read / Write)

concurrency = 128 # The number of concurrent connections to be opened

min_id = 0 # The minimum value for the range of docset_ids generated

max_id = documents # 500 The maximum value for the range of docset_ids generated (when set to the 'documents' value find() executed during read tests will always return 1 document - decrease in order for find() to return multiple documents)

connstr = "mongodb://localhost:28017/" # The default connection string - username and password can be added

str_cnt = 1 # number of UUID string elements in document (increasing this value will increase document size)

```