Scalability - a talk at LugRadio Live 2007
I’m at LRL (currently watching Ted Haeger’s talk), and I am publishing my notes for my talk tomorrow, “Scaling Up for Champions”.
There are quite a lot of config examples in the notes which aren’t on the slides to support the talk.
Overview:
- Definition of Scaling & Problems of scaling
- You have to monitor things
- Scaling individual machines - Disk IO, Processors, Memory, Connectivity
- Multiple Servers - L7 proxying, L4 switching, CDNs
- Memcache
- Simple tuning
Its a 30 minute talk so its an overview of all of those things.
Comments
4 Responses to “Scalability - a talk at LugRadio Live 2007”
Leave a Reply
You must be logged in to post a comment.
July 9th, 2007 at 1:58 pm
That was a great talk :) Thanks for putting it together for us! I may have been very hung over, but I liked it.
Something that particularly caught me was Zabbix, I’ve not seen that before. At the moment I tend to monitor my network using custom PHP agents on the boxes (they all have servers with PHP, why use anything else?) with centralised datalogging on a monitoring box. I might have to give it a go.
Also, throwing in little tweaks is always brilliant: everyone has different tweaks, so sharing them is always good :)
July 14th, 2007 at 1:01 pm
Hello!
thank you for putting the slides online; I did walk into your talk, and listened to the first half which seemed quite obvious stuff, and ended up leaving (I was also on the ‘front door’ at the time so felt guilty about slacking off)….
I think my wife was organising the room at the time, and she mentioned that you covered some cool stuff later on - so I felt a bit annoyed with myself for not sticking around and listening to the rest…
So - thanks for putting the slides online!! :-)
David
( http://codepoets.co.uk )
July 15th, 2007 at 12:35 am
David - thanks for the kind comments. I tried to put something in the talk that would be new to everyone, so you probably saw the sys-admin focussed stuff that was aimed at developers and people wanting to break into sys-admin, and networks people. Later on I go into network geek stuff, that may be new to some sys-admins.
I was nervous about how well it went, so thanks for the feedback from Katherine too.
August 1st, 2007 at 9:33 am
Ey Up Andy
Happy Yorkshire Day by the way :-)
Stumbled on your site recently and been watching it with interest and am enjoying reading these slides, though I have been distracted and wandered off to look at Ultramonkey. Will be giving Zabbix a look soon too, currently thinking about transitioning to Cacti from a rather clunky tool going by the name of SystemStats2 - basically rrdtool and a bunch of commands run over ssh and through awk. Given I have previously seen people mention Zabbix as better than Cacti it might be worth making the jump.