I Tried GridFS in .NET Core… and the Results Were INSANE 🔥
Have you ever struggled with storing large files in your application? Blob storage felt too complex, and a separate file server felt too slow. Recently I decided to test MongoDB GridFS with .NET Core on a real project, and the result was a clean, scalable file-storage solution with no third-party storage and no extra moving parts.
If you want the complete, runnable version of what I describe below - the GridFS service, the upload/download/delete endpoints, and the configuration wired together - it is on Patreon, ready to clone and adapt rather than rebuild from notes.
What Is GridFS and Why It Matters
Unlike traditional file storage, GridFS splits large files into smaller chunks and stores them directly inside your MongoDB database. It also keeps metadata, so you can tag, search, and manage files with ease.
Key benefits:
- Handles files of any size, beyond the 16 MB BSON document limit
- No separate file server required
- Supports metadata and tagging
- Integrates cleanly with existing applications
It is a strong fit for apps that need reliable file handling - PDFs, images, videos, or any large media assets - without standing up dedicated storage infrastructure.
How I Integrated It with .NET Core
Here is the shape of how I approached it in my project:
- Set up MongoDB and configured the connection in the app.
- Installed the MongoDB driver that provides GridFS support.
- Registered the MongoDB client as a singleton service (the driver is thread-safe and designed to be reused).
- Created a lightweight service to handle upload, download, filtering, and delete operations through the GridFS bucket.
- Exposed simple API endpoints to interact with the service from any front end or external system.
No complex pipelines and no expensive cloud storage - just a few deliberate steps to get file storage running inside the database itself.
Real Test Results
Once everything was wired up, I ran a few tests:
- Uploaded image and PDF files successfully
- MongoDB automatically created the
filesandchunkscollections - File downloads were fast and stable
- Tag-based filtering worked exactly as expected
- Deletion was clean and immediate
Performance stayed smooth even with large files, which is exactly what I wanted.
A Small Detail That Makes a Big Difference
GridFS lets you configure chunk size, which can noticeably affect performance depending on your use case. For my demo I reduced the chunk size for better streaming efficiency, but the default 255 KB works well for most scenarios. Tune it only when you have a streaming or memory profile that justifies it.
Why This Combination Works So Well
Using .NET Core with GridFS brings together two complementary strengths:
- .NET Core gives you clean, high-performance backend APIs.
- GridFS provides scalable, metadata-friendly file storage built right into MongoDB.
Together they remove the need for external file servers and simplify your infrastructure - especially useful for SaaS products, dashboards, internal tools, or any system where file handling is a core feature.
Final Thoughts
File storage does not have to be complicated. With GridFS and .NET Core you can build a powerful, flexible, and maintainable file-management solution inside your application, without extra moving parts. It will not replace a CDN for globally distributed media, but for application-owned files that need metadata and search, it is hard to beat for simplicity.
FAQ
What is MongoDB GridFS? GridFS is a specification in MongoDB for storing and retrieving files larger than the 16 MB BSON document limit. It splits each file into chunks stored in a chunks collection and keeps file metadata in a files collection, so you can stream, tag, and query files directly from the database.
When should I use GridFS instead of blob or cloud storage? Reach for GridFS when you want files co-located with your data, need metadata and querying, and want to avoid standing up a separate file server - common for internal tools, dashboards, and SaaS apps. For globally distributed public media at scale, a CDN or object store like S3 is usually the better fit.
What chunk size should I use in GridFS? The default 255 KB works well for most workloads. Reduce it when you need finer-grained streaming, and consider larger chunks for very large sequential files. Measure with your real file sizes and access patterns before tuning.
Do I need a third-party library for GridFS in .NET Core? No. The official MongoDB .NET driver includes GridFS support through a GridFS bucket API, so you can upload, download, and delete files without any extra storage dependency.
About the Author
Celin Daniel is Co-founder of Coding Droplets with 13+ years of hands-on experience building, shipping, and operating .NET and ASP.NET Core systems in production. The guidance here comes from real projects and production incidents, not theory.
- Website: codingdroplets.com
- GitHub: github.com/codingdroplets
- YouTube: youtube.com/@CodingDroplets
