Sometimes bad events can lead to insight. That is what happened to me this week.
Power Failure Results In 4-Day Delay To Final Repair.
My company maintains older hardware that IBM no longer supports. One of my favorite client’s legacy Intel server with Windows 2003 failed on a Saturday afternoon due to power failure that affected their entire building.
This server was flat non-functional. From what little information we could gather remotely, it took one of my top engineers and myself 4 hours the following Sunday to determine that the most likely cause of failure was a system board and power regulator. We also located a source to ship in a complete replacement system for harvest of repair parts to replace other items, in the event of cascading failure.
6:00 a.m. Monday morning we order the replacement server to ship overnight to arrive on site next day Tuesday. Next day, the overnight shipping company FAILS to load the replacement server on the truck…so it finally arrives on Wednesday. UGH!
So what?
It took 4 days to recover.
Result?
Unexpected delays to fix a straight-forward hardware repair.
Insights?
1) Reliable Data Center. Reliable Hardware — 99.999% Uptime
If this application had been hosted, it would be on newer, more reliable standardized hardware at a safe secure data center.
In other words, there would be NO POWER FAILURE with data center power redundancy.
No hardware failure most likely, because the hardware is newer and more reliable. (My top engineer uncovered that the specific Intel server had a history of system board failures. As a hosting provider, we research the reliability of our hardware to avoid these design flaws.)
2) Redundancy And Faster Response To Potential Disruption — 99.999% Uptime
Hosted servers at secure, redundant data centers have greater security and redundancy so such an error should NOT occur.
In the unlikely event a hardware problem does occur, the system redundancy should compensate until the bad part can be fixed or replaced….which should happen within a matter of minutes to about an hour…not 4 days. Parts are stocked on site and at a backup location, so there would NEVER be a delay to overnight parts or have the carrier delay the delivery.
3) Responsive Technical Support Onsite, Online And Minutes To Data Center — 99.999% Up-time
Hosting providers’ technical support is readily available. Depending on the time of day and provider, technical support is on premise, online or minutes away from the hosting data center.
In the unlikely event of a hardware issue, technical support is readily available to deal with it without delay. Not hours or days. No freight carrier delays either.
Conclusion
As I reflected on the events and the unfortunate consequences of this server repair, I concluded that hosting would provide most all of my clients with greater protection, security, and continuity.
I also know first-hand how much faster and easier we can attend to client needs from a team of technical experts in a central location, with hands-on the hosting hardware versus when we provide hardware and software support for remote clients, even with manufacturer’s support.
Even the best manufacturers have delays to get onsite with parts for maintenance.
This is not an issue just for hosted applications.
Now I am even more convinced about the advantages of hosting to provide continuous IT computing with centralized expert support.
Learn in 2 minutes why we are one of IBM’s Premier Business Partners for IBM i Hosting
Leave a Reply