Anatomy of a Read Operation

Let’s learn how GFS handles a read operation.

A typical read interaction with a GFS cluster by a client application goes like this:

1. First, the client translates the file name and byte offset specified by the application into a chunk index within the file. Given the fixed chunk size, this can be computed easily.
2. The client then sends the master an RPC request containing the file name and chunk index.
3. The master replies with the chunk handle and the location of replicas holding the chunk. The client caches this metadata using the file name and chunk-index as the key