> On a local SSD, a random read costs nothing if a page is cached, and costs under a millisecond in case of a page fault. […] The scarcest resource is not CPU, and not bytes, but the number of round trips you make to the object store, and most design decisions below follow from that

There's not really a difference here between S3 and your disk, aside from the literal amount of latency. Restructuring the format to require less reads to make S3 faster would proportionally benefit disk reads.

S3 is a lot easier manage, though. Storage is just then AWS's problem. (And what's a few ms of latency on the BE when pages take tens of seconds to load on the whole.)

Well, it answers the technical questions of what changes when you stick your search indexes on slower and more expensive storage - but not why you would do that in the first place.

Isn't this the same as Elasticsearch searchable snapshots?

Reminds me of turbopuffer