Comment by nine_k
1 day ago
I think these are different use cases. If you talk about SIMD, you talk about the CPU and efficient processing of large numbers of integers. I think that when a solution like this crops up, it's about storage or transmission, and dense packing at the cost of non-uniformity. It's more like time-series databases pack numbers by delta encoding.
The thing is, most real-world numbers will fit within 1-3 bytes (even at 7 bits per byte), so ultradense packing doesn't actually buy much outside of benchmarks.
I spent WAYYYYYYYY too much time exploring this...
I dunno. Varints in the wild tend to be misused, and there are external proto schemas at work we have to integrate with which would literally be both faster and smaller as gzipped json. They're misused because they have an API encouraging misuse -- compressing scalars rather than sequences. Varints are used because they can have reasonable developer ergonomics while sometimes improving computer metrics a twidge.
On top of that, for the vast majority of performance/cost parameter spaces, you're better off both in developer ergonomics and speed/space slapping zstd across a flatter binary format, supposing no better tool fits your use case better. Especially if your messages aren't exceptionally tiny. You're not using them in a raw DB or doing raw bulk analysis on varints (else basically zero choices of parameters make varints win out), so you're transferring them somewhere and decoding them. That decoding step, even for highly optimized solutions like bijou64, is on par with (slightly better than, if you have an older datacenter link) your raw network. If you spend 1s on networking, you spend 1s on parsing. That's a bad tradeoff almost always, and that assumes a good varint solution.
Even when varints make sense for some set of perf/cost parameters, it's still only for developer ergonomics 99.9999% of the time. Even simple changes like operating on a sequence of values rather than a single scalar enable vastly better CPU/space tradeoffs, and being willing to craft a proper data layout usually offers huge gains on top of that.
It's interesting that you pick delta encoding (or, its natural extension, double-delta encoding often being valuable) for time-series databases as an example. That's an obvious case where you have a solution which is extremely cheap in storage/network/CPU. Varints suck comparatively, almost always.
Not to rip on them too much, especially since it's nice to have primitives available which let you not have to do hard thinking for literally every problem, but they're not amazing and not a great default.
Stored and transmitted data also has to be decoded. With modern datacenter hardware bottleneck is often CPU rather than network or disk (SSD). It depends on specific properties of the data. (I used to work on search index implementation which is about decoding and intersecting large amounts of hit-lists; and right SIMD-friendly varint encoding is obviously crucial)