napkin-math VS sccache

Most production storage systems/databases built on top of S3 spend a significant amount of effort building an SSD/memory caching tier to make them performant enough for production (e.g. on top of RocksDB). But it's not easy to keep it in sync with blob...

Even with the cache, the cold query latency lower-bound to S3 is subject to ~50ms roundtrips [0]. To build a performant system, you have to tightly control roundtrips. S3 Express changes that equation dramatically, as S3 Express approaches HDD random read speeds (single-digit ms), so we can build production systems that don't need an SSD cache—just the zero-copy, deserialized in-memory cache.

Many systems will probably continue to have an SSD cache (~100 us random reads), but now MVPs can be built without it, and cold query latency goes down dramatically. That's a big deal

We're currently building a vector database on top of object storage, so this is extremely timely for us... I hope GCS ships this ASAP. [1]

[0]: https://github.com/sirupsen/napkin-math

Trying to estimate performance using some napkin math based on this: https://github.com/sirupsen/napkin-math

So napkin maths time. Typical cross-world bog-standard network speeds for a single TCP channel of ~25MiBps. A single HEADERS+RST pair is likely < 128 bytes (40 for the HEADERS + whatever payload, and 32 for the RST). So 8 pairs per K, 8K pairs per MiB, 200K pairs per 25MiB...

Yes, sequential I/O bandwidth is closing the gap to memory. [1] The I/O pattern to watch out for, and the biggest reason why e.g. databases do careful caching to memory, is that _random_ I/O is still dreadfully slow. I/O bandwidth is brilliant, but latency is still disappointing compared to memory.

[1]: https://github.com/sirupsen/napkin-math

https://github.com/sirupsen/napkin-math (memorize these)

Use icecream or sccache. sccache supports distributed builds.

https://github.com/mozilla/sccache/blob/main/docs/Distribute...

Worth noting that the first commit in sccache git repository was in 2014 (https://github.com/mozilla/sccache/commit/115016e0a83b290dc2...). So I suppose that what "happened" happened waay back.

If you have lots of shared dependencies, maybe try sccache?

I'm going to set up sccache [0] to use it tomorrow. We use MSVC, so EFS is off the cards.

[0] https://github.com/mozilla/sccache/blob/main/docs/S3.md

I think the primary benefit of pre-built procmacros will be for build servers which don't use a persistent cache (like sccache), since they have to compile all dependencies every time. But IMO improved support for persistent caches would be a better investment compared to adding support for pre-built procmacros.

Checkout https://github.com/mozilla/sccache

https://github.com/mozilla/sccache is another option which addresses the use cases of both icecream and ccache (and also supports Rust, and cloud storage of artifacts, if those are useful for you)

That being said a compilation cache, eg the de-facto standard for Rust: sccache (https://github.com/mozilla/sccache) will help to compile and store some of the build artifacts centralized - still for each crate version + build profile (RUSTFLAGS) combination.