Created an issue as you suggested: https://bugs.chromium.org/p/v8/issues/detail?id=13772
On Fri, 24 Feb 2023 at 09:54, Victor Polevoy <[email protected]> wrote: > Thank you for your prompt response. > I am not sure how, but this commit > https://github.com/v8/v8/commit/8c7c087812e343ed8f2c37b21a29aa0de560268d > decreases the speed by 10-20 times. > > On Thu, 23 Feb 2023 at 15:29, Jakob Gruber <[email protected]> wrote: > >> Hi Victor, >> >> please bisect into V8 and if you find the culprit commit, file a bug at >> crbug.com/v8/new. >> >> On Thu, Feb 23, 2023 at 2:25 PM Victor Polevoy <[email protected]> >> wrote: >> >>> Forgot to mention two things. >>> To define this function, the redis-cli has to be used: >>> >>> ❯ redis-cli >>> 127.0.0.1:6379> RG.FUNCTION LOAD "#!js name=lib\n >>> redis.register_function('foo', ()=>{return 1;});" >>> >>> Another thing is that one has to have Redis installed and the RedisGears >>> module loaded. The easiest way to start is the README within the >>> repository: >>> https://github.com/RedisGears/RedisGears/blob/master/README.md#getting-started >>> On Thursday, 23 February 2023 at 14:22:14 UTC+1 Victor Polevoy wrote: >>> >>>> We found a huge performance degradation from version *10.8.168.21* >>>> until version *10.9.194.9*. >>>> >>>> Version 10.8.168.21 >>>> Summary: throughput summary: *469483.56* requests per second latency >>>> summary (msec): avg min p50 p95 p99 max 0.205 0.168 0.183 0.327 0.863 1.055 >>>> >>>> Version 10.9.194.9 >>>> Summary: throughput summary: *20759.81* requests per second latency >>>> summary (msec): avg min p50 p95 p99 max 2.356 0.584 2.383 2.735 3.023 5.999 >>>> >>>> The latest available version, *11.0.226.16*, still has the same issue. >>>> >>>> We use V8 in our RedisGears project ( >>>> https://github.com/RedisGears/RedisGears/). We use redis-benchmark for >>>> performance measurements - if you have Redis installed and running, define >>>> the function: >>>> >>>> *RG.FUNCTION LOAD "#!js name=lib\n redis.register_function('foo', >>>> ()=>{return 1;});"* >>>> >>>> And then run the benchmark: >>>> >>>> *redis-benchmark -c 1 -P 100 rg.fcall lib foo 0* >>>> >>>> We had to downgrade to the 10.8.168.21 version as a decrease of 20x is >>>> a bit too much. >>>> >>>> -- >>> -- >>> v8-dev mailing list >>> [email protected] >>> http://groups.google.com/group/v8-dev >>> --- >>> You received this message because you are subscribed to the Google >>> Groups "v8-dev" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to [email protected]. >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/v8-dev/916ffcfa-bbf2-4616-abf4-481fe804122fn%40googlegroups.com >>> <https://groups.google.com/d/msgid/v8-dev/916ffcfa-bbf2-4616-abf4-481fe804122fn%40googlegroups.com?utm_medium=email&utm_source=footer> >>> . >>> >> -- >> -- >> v8-dev mailing list >> [email protected] >> http://groups.google.com/group/v8-dev >> --- >> You received this message because you are subscribed to a topic in the >> Google Groups "v8-dev" group. >> To unsubscribe from this topic, visit >> https://groups.google.com/d/topic/v8-dev/sVhPu-PSlrc/unsubscribe. >> To unsubscribe from this group and all its topics, send an email to >> [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/v8-dev/CAH3p7oM0JVQj9frnV6s7O4eCsQn-ii_B8XZhObedLVpsSdJuAA%40mail.gmail.com >> <https://groups.google.com/d/msgid/v8-dev/CAH3p7oM0JVQj9frnV6s7O4eCsQn-ii_B8XZhObedLVpsSdJuAA%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> > > > -- > Best, Victor Polevoy. > -- Best, Victor Polevoy. -- -- v8-dev mailing list [email protected] http://groups.google.com/group/v8-dev --- You received this message because you are subscribed to the Google Groups "v8-dev" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/v8-dev/CAMSmOEXD_%3DYgbx90s9obdNcxYFsSyxzyzji5UV%2B5r9pBeG5wBQ%40mail.gmail.com.
