Capacity and Availability Management
February 22nd, 2007Those sound like some “pretty big words” and for most of us it’s about how big’s the pipe and what’s the uptime; but to a number of the clients we are working with it’s about auditing I.T. Infrastructure and provider Service Level Agreements to ensure ongoing operations are meeting their defined criteria.
There are a number of key open source software applications that are critical in providing cost effective solutions to achieve visibility and history of your environment and these are a few we can’t rave enough about!
Capacity Management
If your not familiar with Cacti once it is configured it can provide you with a visual representation of how your infrastructure is performing allowing you to create baselines and trends to allow you to scale or size your infrastructure requirements. There are a ton of templates available from the Cacti community that would get monitoring your different appliances and systems quite quickly.
One of the biggest things I tell our clients is “if you can manually do it” we can figure out a way to turn that into a graph to plot it against your operations.
Availability Management
Nagios or as it was previously know Netsaint is a key application to monitor the health or availability of your environment. With Nagios you can alert based on applications, devices or services not performing to their configured parameters (i.e. failing health checks). Plug-in’s are a dime a dozen and while they are simple to construct there are a large number of them provided by community developers of the Nagios project. There are people out there that are monitoring disk space to ups’s to environmentals.
I sound like a broken record but once again “if you can manually check something not performing” there are ways to automate that and allow your engineers to get back to drinking coffee or reading the newspaper (all the useful things).
Configuration Management
“Change Control” I’m sure almost any engineer would rather just make changes than fill in countless numbers of change forms. With rancid it allows you to audit your Cisco (or other vendor supported) environment to detect changes to platform and configuration to ensure you can tie up any loose ends. While Rancid isn’t designed to be a replacement for your change control procedures it allows you to streamline your process’s to ensure your teams know the state of the environment. With Rancid you have the ability to apply bulk changes via single commands thus bringing down the total cost of management.
Taking the time to invest into proactive management of your environment can pay itself back many, many times over as we have seen time and time again. While the details of the open source applications we have listed above aren’t the be-all and end-all it’s a good place to start. There are many commercial applications in the three defined groups above and many tie closely to the ITIL framework which may or may not be required within your environment.
AdventNet provide some very cost effective commercial implementations that can really tie together some of the dispirit systems you may have laying around inside your environment. Why not have a look on their web site.
We look forward to any questions or suggestions you may have on how people can enhance or provide proactive management to their environments. 