MongoDB Benchmarker

This is a simple tool to benchmark MongoDB performance. It allows bulk updates, inserts, reads, and deletes. It generates random data as you want. For each query, it measures the time it took to execute it and the number of operations per second.

Installing

You need operations.json and operations.log files to use the application properly.

• operations.json file is a configuration file for bulk operations. You have to configure it for use with Docker. However, if you want to use CLI mode, you don't need it.
• operations.log file is the log file that will store benchmark results of the executed operations.

With Clonning Repository

$ git clone git@github.com:hamza-cskn/MongoDBBenchmarker.git
$ cd MongoDBBenchmarker
$ dotnet run

With Docker

Click here to see the Docker Hub page.

Via Compose

Download the docker-compose.yml file. Go to the directory of the file. Configure an operations.json file. There is an example in the repository. Create an operations.log file. (touch operations.log is fine.)

$ docker-compose up

Via Command

Configure an operations.json file. There is an example in the repository. Create an operations.log file. (touch operations.log is fine.)

$ docker run -v <PATH_TO_CONFIG_FILE>:/app/operations.json:ro \
-v <PATH_TO_LOG_FILE>:/app/operations.log \
--rm mongodb-benchmarker

Features

MongoDB Benchmarker supports Insert, Update, Read, and Delete operations.

Logging

Every operation is stored in operations.log file. Here are example logs.

2023-08-23 21:14:12 - INSERT - Count: 1000, Time: 00:00:00.1622390, Template: {name:"saadasdsa"}
2023-08-23 21:14:33 - INSERT - Count: 100000, Time: 00:00:09.7082600, Template: {name:"saadasdsa"}
2023-08-23 21:14:55 - READ - Count: 101000, Time: 00:00:01.1733350, Filter: { "name" : "saadasdsa" }
2023-08-23 21:15:38 - DELETE - Count: 101000, Time 00:00:16.7651800, Filter: { "name" : "saadasdsa" }
2023-08-23 21:15:42 - READ - Count: 0, Time: 00:00:00.4873030, Filter: { "name" : "saadasdsa" }
2023-08-23 21:24:14 - READ - Count: 5, Time: 00:00:00.5765820, Filter: { "name" : "hamza" }
2023-08-23 21:25:07 - READ - Count: 0, Time: 00:00:00.4542770, Filter: { "name" : "hamza" }
2023-08-23 21:26:25 - READ - Count: 0, Time: 00:00:00.7174220, Filter: { "name" : "Jhon", "age" : 20 }
2023-08-23 21:27:00 - READ - Count: 0, Time: 00:00:00.4747580, Filter: { "name" : "Jhon" }
2023-08-23 21:28:11 - INSERT - Count: 3000, Time: 00:00:00.4328780, Template: {name:"Jhon"}
2023-08-23 21:29:11 - UPDATE - Count: 3000, Time 00:00:00.9136250, Filter: { "name" : "Jhon" }, Updated: { "$set" : { "name" : "notjhon" } }

Placeholders

You can specify placeholders in template documents. The document generator will handle them before bulk inserting.

{name: "Hamza", age: "%int(10,80)%", resume: "%string(500)%"}

Placeholder List

• Id: Iteration number.
  • Format: %id%
• Random String randomized string with a specific length.
  • Format: %string(n)%
• Random Integer random integer in a specific range.
  • Format: %int(5,10)%

Config Mode

Use dotnet run --config command to get inputs from the operations.json file.

In config mode, no input will be received from the user. So, you have to specify connection credentials using these environment variables:

• BENCHMARK_DATABASE_NAME
• BENCHMARK_COLLECTION_NAME
• BENCHMARK_CONNECTION_STRING

Also, you can specify a location of operations.json file with BENCHMARK_CONFIG_PATH variable. It is completely optional.