System downtimes are unavoidable. After all, a computer system is comprised of a multitude of electrical parts that can fail. The risk of failure and the recovery time from failure could be reduced significantly if the following recommendations were put in place.
1. Increase support contract from 16x6 to 24x7. If 24x7 is overly costly, then have a standing purchase order in place with the service provider for after hour support that includes payment for time and material.
2. Replace existing system with an enterprise class system that does not include any single points of failure. An example would be redundant disk controllers, redundant I/O boats, redundant CPU/memory modules, etc. Redundancy doesn’t necessary mean 100% uptime, but if a module goes down the system can be failed over to the remaining module until a replacement arrives.
3. Power to the system would include commercial and locally generated power including enterprise class UPS.
4. Redundant network connection to NERO/OWEN.
5. During downtimes, NEVER GIVE UP. Work the issues until the system is up in production again.
Leave a comment