Comments on: Context and Operational Metrics http://www.kitchensoap.com/2009/05/10/context-and-operational-metrics/ Thoughts on capacity planning and web operations. Fri, 19 Feb 2010 11:07:18 -0800 http://wordpress.org/?v=2.8.6 hourly 1 By: DevOps Cooperation Doesn’t Just Happen With Deployment http://www.kitchensoap.com/2009/05/10/context-and-operational-metrics/comment-page-1/#comment-7836 DevOps Cooperation Doesn’t Just Happen With Deployment Wed, 16 Dec 2009 01:28:26 +0000 http://www.kitchensoap.com/?p=195#comment-7836 [...] said this before, but context is absolutely everything. Application-level or feature-level metrics is what gives the missing context to in-the-box [...] [...] said this before, but context is absolutely everything. Application-level or feature-level metrics is what gives the missing context to in-the-box [...]

]]> By: JIRA: Development Infrastructure http://www.kitchensoap.com/2009/05/10/context-and-operational-metrics/comment-page-1/#comment-7580 JIRA: Development Infrastructure Tue, 02 Jun 2009 15:03:30 +0000 http://www.kitchensoap.com/?p=195#comment-7580 <strong>[DEV-52] Operations Management > Essential Monitoring...</strong> As someone who's interested in how things are working on a particular machine, it might be useful to determine how to effectively measure the health of a specific machine. For example, without getting corny is there a 10-point (really N-point) measure... [DEV-52] Operations Management > Essential Monitoring…

As someone who’s interested in how things are working on a particular machine, it might be useful to determine how to effectively measure the health of a specific machine. For example, without getting corny is there a 10-point (really N-point) measure…

]]> By: William Louth http://www.kitchensoap.com/2009/05/10/context-and-operational-metrics/comment-page-1/#comment-7537 William Louth Fri, 15 May 2009 15:14:02 +0000 http://www.kitchensoap.com/?p=195#comment-7537 I think you can go even further than just offering contextual counters but also correlating the rate of change in such counters with the context of the activity chain which activities associated with one or more cost groups (centers) that can represent the business domain, implementation codebase, or user activity. You might find the following links useful - as low or as high as you want to go. Execution Profiling: Counting KPI’s http://williamlouth.wordpress.com/2009/05/04/execution-profiling-counting-kpis/ ABC for Cloud Computing http://williamlouth.wordpress.com/2009/01/27/abc-for-cloud-computing/ I think you can go even further than just offering contextual counters but also correlating the rate of change in such counters with the context of the activity chain which activities associated with one or more cost groups (centers) that can represent the business domain, implementation codebase, or user activity.

You might find the following links useful – as low or as high as you want to go.

Execution Profiling: Counting KPI’s
http://williamlouth.wordpress.com/2009/05/04/execution-profiling-counting-kpis/

ABC for Cloud Computing
http://williamlouth.wordpress.com/2009/01/27/abc-for-cloud-computing/

]]> By: Javier http://www.kitchensoap.com/2009/05/10/context-and-operational-metrics/comment-page-1/#comment-7529 Javier Mon, 11 May 2009 02:49:08 +0000 http://www.kitchensoap.com/?p=195#comment-7529 Great post, John! I think there's other examples of context enabling metrics that are not necessarily based on relationships between traditional ops metrics (as your example suggests). For example, we've worked with folks who track the number of photo uploads, or new subscribers alongside all the rest of their ops metrics and events. These "business" metrics are understood by both ops and the business folks, and at sufficiently mature sites behave in fairly consistent patterns. A lot of our customers start by tracking anomalies in these metrics and work their way down from there to diagnosing the problem. Exposing this context is something developers should have an active hand in and too often are not. I'm eager to learn ways in which dev and ops are working together to make this more a reality for anyone doing scaled out webapps. See you at Velocity. -javier Great post, John!

I think there’s other examples of context enabling metrics that are not necessarily based on relationships between traditional ops metrics (as your example suggests). For example, we’ve worked with folks who track the number of photo uploads, or new subscribers alongside all the rest of their ops metrics and events. These “business” metrics are understood by both ops and the business folks, and at sufficiently mature sites behave in fairly consistent patterns. A lot of our customers start by tracking anomalies in these metrics and work their way down from there to diagnosing the problem.

Exposing this context is something developers should have an active hand in and too often are not. I’m eager to learn ways in which dev and ops are working together to make this more a reality for anyone doing scaled out webapps.

See you at Velocity.

-javier

]]>