Scalability Boot Camp

SXSW 2008. A panel talk given with several other startups discussing how to prepare for scalability challenges. Moderated by Kerry Miller of BusinessWeek.

Transcript

1. Scalability Boot Camp SXSW 2008 Jakob Heuser Alan Kasindorf Blaine Cook Sandy Jen Kerry Miller
2. Briefing Panelists Kerry Miller - BusinessWeek Alan Kasindorf (aka dormando) - Six Apart Jakob Heuser - Gaia Online Blaine Cook - Twitter Sandy Jen - Meebo Kerry Miller - BusinessWeek
3. Why Scale • Good problem • Doesn’t have to cost • It’s an “everybody” thing Kerry Miller - BusinessWeek
4. The Regimen • Problem • Concepts • Abstract Solutions • Google Time • Conversation Kerry Miller - BusinessWeek
5. Monitoring avoid working in the dark Sandy Jen - Meebo
6. Get it on the radar • Understand the “pain points” • Live and die by monitoring • Monitor EVERYTHING Sandy Jen - Meebo
7. “Everything”? • Disk I/O • Memory • Bandwidth • Page Load Times • The list goes on... Sandy Jen - Meebo
8. Google Time • Ganglia • Hyperic • sar and sysstat - simple, you already have it • Know your tool, whatever it is Sandy Jen - Meebo
9. File System Solutions users like making tons of lolcats and storing them on your website Jakob Heuser - Gaia Online
10. Never gonna be a hero Disk IO Graph Jakob Heuser - Gaia Online
11. Use what you have • Don’t waste capacity • Use someone else’s space • Avoid a single “authority” on a file Jakob Heuser - Gaia Online
12. Google Time • DRBD + OCFS • Amazon S3 • MogileFS (Danga Software) • lustre Jakob Heuser - Gaia Online
13. The Database Layer your most common, but hardest to solve bottleneck Alan Kasindorf - Six Apart
14. Enemy diversion Disk IO CPU Usage Alan Kasindorf - Six Apart
15. The real problem Show Processlist Output Alan Kasindorf - Six Apart
16. Another war zone • Make “:) SQL” not “:( SQL” • Horizontal Partitioning • Caching Layer Alan Kasindorf - Six Apart
17. Google Time • Memcache • HiveDB • CouchDB / Hypertable • MySQL Consultant Alan Kasindorf - Six Apart
18. Parallel Processing you don’t have to do it all right now Blaine Cook - Twitter
19. Under siege Slow Query Log Blaine Cook - Twitter
20. Smarter, not stronger • Consistent for current user, not everyone • Design code for parallel steps • cronjobs Blaine Cook - Twitter
21. Google Time • Starling • Gearman • TheSchwartz Blaine Cook - Twitter
22. Regroup Jakob Heuser - Gaia Online
23. Regroup • All these technologies are built to be asynchronous • An amazing amount of your app can be asynchronous too Jakob Heuser - Gaia Online
24. Scalability Boot Camp SXSW 2008 Jakob Heuser Alan Kasindorf Blaine Cook Sandy Jen Kerry Miller http://www.slideshare.net/Jakobo
25. Resources Monitoring http://ganglia.sourceforge.net http://www.hyperic.com http://www.nagios.org http://pagesperso-orange.fr/sebastien.godard/ http://www.cacti.net CDN http://en.wikipedia.org/wiki/Content_Delivery_Network http://en.wikipedia.org/wiki/Akamai_Technologies
26. Resources File Systems http://www.danga.com/mogilefs/ http://www.lustre.org http://www.drbd.org http://oss.oracle.com/projects/ocfs/ Database http://www.planetmysql.com http://datacharmer.blogspot.com/ http://www.danga.com/memcached/ Job and Queue Systems http://www.danga.com/gearman/ http://rubyforge.org/projects/starling/
27. Image Credit http://www.history.noaa.gov/stories_tales/radar.html (radar) http://www.thatpoliticalblog.com/serendipity/plugin/tag/TPB+Information (bar chart) http://www.digitalearth.com.au/category/general/ (disk on fire) http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=160210671355 (dump truck) http://420.thrashbarg.net/ (marching penguins)