This article is a work in progress. This is based on a mind map of sorts that I've been recently putting together. You can find that here.

Context: system reliability

Summary:

A complex class of failures that feed and strengthen their own fail condition through a positive feedback look that keeps the system in a degraded/failed state.

For example, using a cache to improve the response times of a system (and hence, the user experience), is a great idea. Using a cache to improve the scalability of a system, not so much. When the cache is unavailable, the system, that's not designed to cope with the amount of traffic, will start failing. Clients attempting to retry their requests will send the system into a metastable failure, not allowing it to recover.

Follow-up: start around here and have a go around the mind map.

Context: diagnosing slow-performing Java apps

Follow-up: start around here on the mind map.

… coming soon: this is something I'm interested in but haven't yet gotten to have a play with.

Read more about it here.

… coming soon: this is something I'm interested in but haven't yet gotten to have a play with.

Read more about it here.