notes from LWN about RPS patch — a way to distribute incoming traffic coming from 1 source (1 NIC, 1 queue, 1 IRQ line) via multiple CPU by hashing tcp headers and distributing by hash across multiple CPU/Cores
Posted in c, for:collidr, google, kernel, Linux, network, patch, performance, programming | Comments Off | permalink
read this before rushing to by some SSDs
Posted in caching, database, disk, for:collidr, IO, performance, SSD, sysadmin | Comments Off | permalink
also see http://dormando.livejournal.com/521163.html Describing some problems with performance/scalability you may have if you partition/shard your data by hash. ( that’s what memcached does.) You may hit the cpu bound limit from the amount of requ…
Posted in architecture, facebook, high_load, memcache, memcached, multiget, performance, scalability, scaling, system | Comments Off | permalink
test results for linux routing performance on recent intel “nehalem” CPUs and 10G ethernet
Posted in benchmark, Linux, performance, router, routing, test | Comments Off | permalink
tells you about a plugin (a perl script ) for munin that will collect linux disk IO stats and send it to munin for pretty graphs.
Posted in disk, IO, Linux, monitoring, munin, performance, sysadmin, system | Comments Off | permalink
explains use linux IO scheulers with presence of smart RAID controllers.
Posted in database, for:collidr, IO, kernel, Linux, mysql, performance, postgresql, raid, sysadmin, tuning | Comments Off | permalink
very good presentation slides explaining nginx internals, such as main architecture, main event loops, core modules, processing pipelines, etc. Also there is a video of the talk , but it is mandarin, so YMMV 
Posted in architecture, c, development, high_load, internals, modules, nginx, performance, programming, web | Comments Off | permalink
slides from my.opera.com architecture presentation. Iptables/linux lvs -> apche/mod_perl -> mysql replication, varnish, memcached
Posted in iptables, lvs, memcached, mod_perl, mysql, performance, scalability, web2.0 | Comments Off | permalink
high-load web notification system or how to do persistent connections in ajax/js. Useful for counters, friend-status updates and so on. Example is given for libevent-based daemon in perl.
Posted in AJAX, connections, daemon, development, jabber, javascript, libevent, library, performance, Perl, programming, scalability, web2.0 | Comments Off | permalink
seems to a rare thing: a message queue server, running as a native code
Posted in asynchronous, development, distributed, messaging, performance, Perl, programming, queue, ruby, scalability, server, web2.0 | Comments Off | permalink
You are currently browsing the archives for the performance category.
Konstantin Antselovich (c) 2004-2011
powered by WordPress
Entries (RSS)
and Comments (RSS).