Astricon 2012 in Atlanta, GA was a jam-packed event, filled with industry professionals and experts. As techs crowded the floors, even more joined the open discussions and best practices meetings.
As one of the first sessions held, “Best Practices for Reliable Carrier Grade Telephony,” the AVOXI team was eager to participate. The discussion mainly focused on reliability engineering and how it relates to safety engineering. In fact, the two are quite interconnected seeing as they are somewhat dependent on one another. The difference between the how, however, is that reliability engineering focuses primarily on costs associated with system downtime, personnel and warranty expenses, and cost of repair; the main focus of safety engineering is not that of costs, rather the preservation of system life against dangerous system failure modes.
The discussion highlights the following on ways to avoid system risks or dangers:
- Delegate safe practices that make sense for your business
- Implement processes that are standardized and somewhat formulated
- Duplicate all important functions
- Run back-ups on a daily, weekly, and monthly basis
- Designate employees with effective de-bugging skills to be ready in case of failures
- Monitor machines, services, resources frequently and consistently
It is also worthy to remember that the main reason for outages occur due to untested changes, hard disks filling up, power and network outages, or servers with compromised security.
Jared Smith and AVOXI’s Pablo Valenciano at Astricon 2012. Jared Smith is currently lead of Fedora Project. He is also the co-writer of “Asterisk: The Future of Telephony.”