-
Notifications
You must be signed in to change notification settings - Fork 643
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Clarify results #281
Clarify results #281
Conversation
@OvermindDL1 Should not we use percentile, instead of average latency to rank by |
50th percentile, 99.99th percentile, and standard deviation would be the best to show as it shows the latency range and how it grows. |
@OvermindDL1 my question is want metric (one ?) use to order is it accurate to differentiate both table ? |
I'd include all information on both tables, but I'd use two table, one sorted on If only Github had the ability to mark a table as sort-by-click-on-header. Could always make a dedicated HTML for that though (github pages or so). |
Don't you think tables will be huge ? My final purpose is to compute result, store them in a @tbrand @OvermindDL1 What do you think ? |
Eh, the tables would be long sure, but that seems fine to me (especially if you can clamp the headers to the top as it scrolls). |
@OvermindDL1 On a race between
I suggest then to keep |
Low latencies at high percentiles most likely means they generate little to no garbage in comparison. :-) |
@OvermindDL1 not understood |
Could you remove the column which is not related to each ranking? Other things are acceptable for me 👍 |
402ddfe
to
dcad804
Compare
* add restify (node) * add restify to node targets on makefile * add dockerfile for restify (node)
* add vibora this code has been tested and it requires `SO_REUSEPORT` on Ubuntu 16.04, therefore can't guarantee anything about the code. Use it at your own risk. * Update benchmarker.cr * update for vibora * fix routing -> type not required * add vibora * use gunicorn for sanic * use http://meinheld.org/ to enable async networking with django * reuse port on tornado * use meinheld / SO_REUSEPORT on django * use meinheld / SO_REUSEPORT on flask
* add vibora this code has been tested and it requires `SO_REUSEPORT` on Ubuntu 16.04, therefore can't guarantee anything about the code. Use it at your own risk. * Update benchmarker.cr * update for vibora * fix routing -> type not required * add vibora * use gunicorn for sanic * use http://meinheld.org/ to enable async networking with django * reuse port on tornado * use meinheld / SO_REUSEPORT on django * use meinheld / SO_REUSEPORT on flask
I've done a little change to the We need to refactor using an other key/value store (to avoid adding dependency) between : |
As long as it's ran via docker or so and not needed to be installed. Why not just use leveldb/rockdb if it's just a KV store needed or sqlite for querying? Or just map? |
@OvermindDL1 I mean a |
Definitely no point in using something as crazy-heavy as redis, just embed leveldb or rocksdb, I use both rather excessively, and they will far far outperform redis. |
@OvermindDL1 sure, it was I have said 😛 'cause it was late last night (and I was not very productive), I take a random tool -> I'll rewrite this part |
@waghanza yep, 50% makes sense, for someone else then 99% may make sense, etc... etc... I really wish github had sortable column by headers, maybe need to set up a github pages so they become clickable. ^.^; |
@OvermindDL1 I have find that it's possible, but very hard to use / maintain (sortable table). my idea is to push results is |
No need for BigQuery, just generate an html file via a template and use some progressively enhancing javascript to make the table sortable via column. |
@OvermindDL1 I know it could be done like that, but I find it mor flexible to separate compuation from layout e.g -> use moreover, compute and generate markdown on |
@tbrand Just after merging this, I'll release 💯 |
* fix jester * dockerfile refactor (jester)
* fix jester, seeing tfb implementation * restore nimble file * fix version
* restore hapijs * add hapi to make targets
* remove unused files (mofuw) * homogenize implementation (inspiring from techempower)
* update nim/mofuw * fixed nest.nim not found
* add Java/act framework * validate working for each framework on travis * remove unused files * Add restify (node) (#276) * add restify (node) * add restify to node targets on makefile * add dockerfile for restify (node) * Introducing Vibora (#285) * add vibora this code has been tested and it requires `SO_REUSEPORT` on Ubuntu 16.04, therefore can't guarantee anything about the code. Use it at your own risk. * Update benchmarker.cr * update for vibora * fix routing -> type not required * add vibora * use gunicorn for sanic * use http://meinheld.org/ to enable async networking with django * reuse port on tornado * use meinheld / SO_REUSEPORT on django * use meinheld / SO_REUSEPORT on flask * Introducing Vibora (#285) * add vibora this code has been tested and it requires `SO_REUSEPORT` on Ubuntu 16.04, therefore can't guarantee anything about the code. Use it at your own risk. * Update benchmarker.cr * update for vibora * fix routing -> type not required * add vibora * use gunicorn for sanic * use http://meinheld.org/ to enable async networking with django * reuse port on tornado * use meinheld / SO_REUSEPORT on django * use meinheld / SO_REUSEPORT on flask * update nim/mofuw (#268) * update nim/mofuw * fixed nest.nim not found * cleanup ranking, display only two tables * format percentile * rank by 99th percentile, for latency * update mofuw (#294) * fix jester (#297) * fix jester * dockerfile refactor (jester) * update tornado to 5.1.0 (#298) * use a redis to store results * fix server_nim_mofuw (#301) * add redis on ci (storage will change) * create a sample resultset * add more details on results tables * Fix jester (#305) * fix jester, seeing tfb implementation * restore nimble file * fix version * use an internal in-memory storage for results * Restore hapijs (#306) * restore hapijs * add hapi to make targets * clarified tiles * compute results from each endpoints (previously was only on GET:/) * restore 50th percentile ranking for latency * Optimize mofuw implementation (#304) * remove unused files (mofuw) * homogenize implementation (inspiring from techempower) * add results sets for each route * remove duplicate target in makefile * fix python version for vibora
Hi,
Following the #238, it COULD be
awesome
to report :req/s
latency
It also COULD help
clarify
results. May developers argue that comparing so languages such asnode
and some such aselixir
(which targets are completely different) are not very relevant.This
PR
show results order byreq/s
AND bylatency
Regards,