Re: REST impact paper
"J. Andrew Rogers" <[email protected]>
| Newsgroups | gmane.culture.people.rohit-khare |
|---|---|
| Message-ID | <[email protected]> |
> On Sep 9, 2017, at 12:38 PM, Stephen D. Williams <[email protected]> wrote: >> We ended up implementing our own highly-optimized JSON parser written in C++11. It was much faster than the open source ones it replaced. Nothing fancy, just tight engineering that prioritized performance. We went back later and reimplemented certain hot paths in assembly using SSE4 for systems that supported it, for another 2.5x improvement in throughput as I recall. > > Github repo or it didn't happen. (I have my own queue of things to publish…) Heh, not my code. I neither wrote it nor own it. These days I have an even better optimization for JSON parsing that you are free to use: don’t allow people to use JSON for data at scale. Unfortunately, that wasn’t an option when that high-performance JSON parser was written. Andrew _______________________________________________ FoRK mailing list http://xent.com/mailman/listinfo/fork