ImageSharp seems to retain ~300-400 MB of managed memory even after disposing all my images. Is this a memory leak?
By default, ImageSharp uses ArrayPool-s for performance reasons, however this behavior is fully configurable. All large buffers are managed by the MemoryAllocator implementation associated to Configuration-s MemoryAllocator property. We are using ArrayPoolMemoryAllocator by default, in order to utilize the benefits of array pooling:
- Less pressure on GC, because buffers are being reused most of the time
- Reduced LOH fragmentation
- When working with unclean buffers is acceptable, we can spare on GC-s array cleaning behavior too
Summary: pooling helps us to reduce CPU work and increase throughput for the cost of a larger memory footprint.
Working in memory constrained environments
Sometimes having ~300 MB memory footprint is not an option. Let's mention a few cases:
- When horizontal scaling is achieved by having multiple memory constrained containers in a cloud environment.
- Mobile applications.
Before scaling down pooling behavior because of unwanted
OutOfMemoryException-s in a cloud or desktop environment:
- Keep in mind that image processing is a memory intensive application! This may affect your scaling strategy. We don't recommend using containers with 1 GB or smaller memory limit!
- Make sure that you are running your service in a 64 bit process!
There are several pre-defined factory methods to create an ArrayPoolMemoryAllocator instance for memory constrained environments. For example ArrayPoolMemoryAllocator.CreateWithModeratePooling might be suitable in most constrained situations:
Configuration.Default.MemoryAllocator = ArrayPoolMemoryAllocator.CreateWithModeratePooling();
Of course, you may also configure a MemoryAllocator on your own Configuration instance.
You can find benchmark results in the original PR which may help to select you a configuration, but they are bit outdated, because our throughput got better since then!
Releasing pools programatically
If your application uses ImageSharp sporadically (eg. generating some images on startup, or on other non-frequent use-cases), you may want to release the retained pools using ReleaseRetainedResources:
Using multiple MemoryAllocator instances in the same process
You need to create and maintain your own Configuration instance, setting a specific MemoryAllocator on it. It's possible to pass custom Configuration instances to methods accross our whole API.