Viewing profile — krenoten

In practice this is the source of a tremendous number of bugs when async tasks don't actually clean up shared state in their Drop impls.

I haven't mentioned async Rust on HN since 2015.

The best solution is to use a 0-copy format like flatbuffers, google's zerocopy library, or serde's borrow attributes https://serde.rs/lifetimes.html#borrowing-data-in-a-derived-..…

only narcs use windows for database workloads :P

yeah, le-be conversions are not generally measurable above noise compared to other database-related work. sled stores arbitrary bytes. endianness is the concern of the person who w…

On sled 0.31 the memory growth does not run away as with the earlier version. Note that some in-memory metadata tracking does accrue as nodes split and sled tracks where to find th…

Just use a hashmap. Hashmaps in some cases can be like 8 orders of magnitude faster than Redis.

I make a lot of money fixing problems in storage layers that these kinds of ideas create. Thanks :]

There is zero reason behind this assumption.

It means pay more attention to reliability than pop infrastructure and internet companies (who can offset poor reliability with human attention or intentionally deprioritize it to …

This was my interpretation as well. I'm going to compare a disk-backed bwtree with a disk-backed ART, both backed by the same pagecache, and maybe end up with an ART that scatters …

Users can rely on sequential recovery. At some point I'll probably write a partial recovery tool that gives you all versions of all keys that are at all present anywhere in the rea…

Indeed. This is why I aggressively checksum everything and pay particular attention to throwing away all data that was written after any detected corruption during recovery. This i…

Yeah, I'm curious about using sled as a more ssd friendly storage engine for mentat. I'm just starting to experiment with datalog implementations, but I think by having harmony bet…

It might not. But the critiques of bw trees in terms of performance that I've seen have not had compelling data in terms of things that matter outside of academia or benchmarking s…

This is honestly a use case I'm experimenting with using a mix of CRDTs and OT. Our systems are becoming more and more location agnostic and I don't feel that our current data infr…

Use it for large scale HTAP! It's great for its flexible use cases at high scales :)

I am a total devotee to their approach to building simulable systems, although I seek to push it even farther and integrate lineage driven fault injection from an early stage. I se…

It's not needed for the single-key atomic record store, which is the sled bwtree index that is the current highest level module. MVCC is implemented in most popular embedded DBs be…

It's modular, and there is a paxos implementation, but it has been built totally in simulation so far and I haven't plugged it into an io layer yet. But this is trivial. That said,…

Currently it's even more basic. The current usable parts are a pagecache following the llama approach, some great testing utility libraries, and an index (that you can use as a kv)…

ALICE showed that's not always true with sqlite. Sled is being built with an extreme bias toward reliability over features, but as the readme says, it has some time to go before re…

SPIN is a nice tool for modeling them. You can extract code from a coq model. You can go into the world of dynamic instrumentation to try to verify invariants in implementations.