GridFS with GoLang (Part 1)

After reading multiple related blogs, I was able to deduce the following plan for my server:

1. Build HTTP server & respective routes using net/http package.
2. The server will initiate a connection with MongoDB to perform CRUD operations.
3. But, to bypass the 16 MB fileSize limit of MongoDB, I had to take the GridFS feature into consideration.

Though GridFS eases the management of files, but, GridFS with Go was a learning curve for me.

While I am writing this, there are really few documentations that could help with all the CRUD operations. Apparently, I did manage to meet the requirements using GridFS with go, and hence, here I am to share my journey.. 😉

➤ What is GridFS ??

GridFS basically is a specification that allows to manage the files with size above 16 MB. As we all must be aware that MongoDB stores data records as “BSON documents” inside a collection.

So, instead of storing the complete file as a single document, which might lead to a document of size above 16MB, GridFS

• divides the files into parts (termed as chunks),
• and stores each chunk as a seperate document.

In addition to this file’s content management, GridFS also takes care of managing the metadata of each file. This is done via two different collections; i.e., fs.Chunks and fs.Files

Now that we have discussed the basic components I’ll be using to build my server, let’s move to the next part to dive into the implementations.