Design file storage and sync
A file sync service looks like an upload form with a folder tree, but the grade comes from three decisions underneath it. Splitting metadata from bytes, cutting files into content-addressed chunks so an edit uploads only what changed, and a per-folder change journal that lets every device catch up from a cursor without ever scanning the tree.
What you'll be able to do
- Separate the metadata service from the blob store and say which one each request touches
- Show with a runnable cell why fixed-size chunks fail on an insertion and content-defined chunks do not
- Design a commit protocol where a file revision can never point at a block that was not uploaded
- Sync many devices from a per-namespace change journal and a cursor, and resolve a concurrent edit with a conflicted copy
Before this: blob-storage-and-cdns, real-time-updates
Log in to read the rest of this lesson
A file sync service looks like an upload form with a folder tree, but the grade comes from three decisions underneath it. Splitting metadata from bytes, cutting files into content-addressed chunks so an edit uploads only what changed, and a per-folder change journal that lets every device catch up from a cursor without ever scanning the tree. This walkthrough runs about 27 minutes, with runnable code you can edit and re-run as you read. A new account gets 7 days of Pro free, no card required.
Log in to continueThe first module of every track is free to read on the web: see what's open in System design.