Hacker News Reader

Comment by latch | Hacker News Reader

I've been curious how they handle indexing given that they allow ad-hoc queries. In the feature list they say that they provide "smart indexing". I wonder if they simply log all queries and generate sparse-indexes (they use MongoDB) on the fly? I wonder if "file storage" counts index space?

View on HN

"File storage" actually only counts large files that are stored with PFFile / ParseFile. It does not include data stored in objects or index space; those are all covered by pricing for API requests.

We do track queries and accesses to create appropriate indexes on the fly. Whenever we used databases in the past, indexes always seemed like something that should be done intelligently out of the box, so we decided to make Parse work that way.

Ok, that makes sense. At the beta pricing I was looking at the free tier and thinking it would be limiting some one to 50k users (IIRC PFUser is a subclass of the PFObject). So now we are limited by API calls, I suppose that would mean inactive users don't really count towards usage, which was a problem I was thinking about.

One question I have is what is the advantage of using PFFile instead of storing in S3?

We think PFFile is a lot easier to use from mobile devices, with the iOS and Android SDKs. A lot of people are actually just using PFFile because they find it simpler than using S3. That said, you can use S3 and Parse in the same application, so they work together fine if you'd rather implement things that way.

I'm currently using Parse for data and S3 for file storage. I will give it a try to PFFile in another project to see how it works.

Reply on news.ycombinator.com