addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscontroller-playcredit-cardcrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobe--smallglobegmailgooglegroupshelp-with-circleimageimagesinstagramFill 1launch-new-window--smalllight-bulblinklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1ShapeoutlookpersonJoin Group on CardStartprice-ribbonprintShapeShapeShapeShapeImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruserwarningyahoo

Erlang LA User Group Message Board › Topic information updated: Erlang.LA next Tuesday!

Topic information updated: Erlang.LA next Tuesday!

A former member

Jay Nelson, Concurrently Serial Erlangist, DuoMark

Topic: Erlang Performance, Understanding the cost scale of Erlang operations.



The old saying about Lisp programmers ("they know the value of everything but the cost of nothing") may hold true for a lot of erlang programmers today. You've been programming erlang for a while, can you answer the following questions?



- What are the erlang operations from fastest to slowest?

- How significant are differences in native data structure access speeds?

- How fast is message passing and can code order modify the speed?

- What's the tradeoff between selective receive and more processes?



Programmers should have a handle on the order of magnitude costs of different language features. This talk discusses a github project I use for doodling in erlang (github.com/jaynel/dooderls­) to discover the answer to these types of questions. I hope to develop a relative scale of performance benchmarking tool from these exercises.



This is very much a work in progress, with the goal of providing a chart of the relative performance of different operations from 10e-9 seconds to 1 second. Having such a tool allows the comparison of new OTP release performance to previous performance so that an engineer's intuition for architectural tradeoffs can stay current with the actual characteristics of the underlying language.



In addition the talk will introduce a full rebar-packaged system with embedded yaws running as an included application under the root supervisor, with integration of dynamic JS libraries and HTML5 for interactive display of results in a web browser.



Powered by mvnForum

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy