How do you configure and tune the WiredTiger storage engine and the cache size for your MongoDB deployment?

1 What is WiredTiger?

WiredTiger is the default storage engine for MongoDB since version 3.2. It is designed to support high concurrency and scalability, as well as compression and encryption. WiredTiger uses a write-ahead log (WAL) to ensure durability and a cache to store frequently accessed data in memory. The cache size and the WAL settings are two important factors that affect the performance of WiredTiger and MongoDB.

2 How to set the cache size?

The cache size determines how much data WiredTiger can keep in memory, which reduces disk I/O and improves query speed. However, setting the cache size too high or too low can have negative consequences. If the cache size is too high, WiredTiger may compete with other processes for memory resources and cause swapping or out-of-memory errors. If the cache size is too low, WiredTiger may evict data from the cache too frequently and cause more disk I/O and latency. The optimal cache size depends on your workload and hardware, but a general rule of thumb is to allocate 50% of the available RAM to WiredTiger. You can set the cache size using the --wiredTigerCacheSizeGB option when starting MongoDB, or by editing the storage.wiredTiger.engineConfig.cacheSizeGB parameter in the configuration file.

3 How to tune the WAL settings?

The WAL is a file that records all the changes made to the data and the indexes by WiredTiger. It is used to recover the data in case of a crash or a power failure. The WAL settings affect how often WiredTiger flushes the data from the cache to the disk and how much disk space is used by the WAL files. You can tune the WAL settings using the --syncdelay option when starting MongoDB, or by editing the storage.syncPeriodSecs parameter in the configuration file. The default value is 60 seconds, which means that WiredTiger flushes the data to the disk every 60 seconds. You can reduce this value to increase the durability of the data, but this may also increase the disk I/O and the CPU usage. You can also limit the size of the WAL files using the --wiredTigerJournalCompressor option when starting MongoDB, or by editing the storage.wiredTiger.engineConfig.journalCompressor parameter in the configuration file. The default value is snappy, which is a fast and efficient compression algorithm. You can change this value to zlib, zstd, or none, depending on your trade-off between compression ratio and CPU usage.

4 How to monitor the performance?

To monitor the performance of your MongoDB deployment, you can use various tools and metrics provided by MongoDB. One of the most useful tools is the mongostat command, which displays real-time statistics about the server status, such as the number of operations, the cache usage, the disk I/O, and the network traffic. You can also use the db.serverStatus() method in the mongo shell, which returns a document with detailed information about the server state, including the WiredTiger statistics. Another tool is the MongoDB Atlas Performance Advisor, which is a cloud-based service that analyzes your queries and indexes and provides recommendations to optimize your performance.

5 How to troubleshoot the issues?

If you encounter any issues with your MongoDB deployment, such as slow queries, high CPU usage, or memory errors, you can use various methods to troubleshoot them. One of the most common methods is to use the explain() method in the mongo shell, which returns a document with the execution plan and the execution statistics of a query. You can use this method to identify the indexes used, the stages involved, and the time spent by each stage. You can also use the profiler and the logs to track and analyze the slow operations and the errors. You can enable the profiler using the db.setProfilingLevel() method in the mongo shell, and you can view the logs using the mongod.log file or the db.getLog() method in the mongo shell.