A public forum for discussing the design of software, from the user interface to the code architecture. Now closed.
I'll be launching a web service in the fall and one major concern I have is generating baseline numbers from real use in the field in order to scale the system later on.
At the moment, I'm thinking of having two monitors - one internal and one external - and then correlating the data.
The internal tool would monitor standard statistics such as CPU utilization, RAM utilization, file system accesses, web server hits, and database server query information.
The external tool would would be an automated test harness logging into a dedicated benchmarking account and attempting the common operations throughout the day. Success/failure and time-to-respond to these requests would be collected.
That's my idea anyway - I have two questions:
a) What tools should I use to collect the data? - for the external part, I'm thinking OpenSTA, but I have no ideas for the internal part
b) Is there a better way to gauge how your service deals with load?
I've been using MRTG with SNMP (for server stats) and Webalizer (for website state) for years without a single hitch, but maybe there are better tools around these days.
Saturday, July 16, 2005
This topic is archived. No further replies will be accepted.Other recent topics
Powered by FogBugz