Web Server Uptime Availability

My web server is maintained by a web host provider. I am presently using a shared server which is a reasonable trade-off between cost and performance. In due time, as traffic volumes dictate, I will seek a virtual server with more dedicated resources (less shared resources). During the week of November 8th, 2010, the server had hardware problems with a hard drive unit. Maintenance was scheduled for Nov 10th to replace the hard drive. During the few days leading up to the maintenance window, the server encountered delays in serving web pages i.e. web pages were slow to load. During the maintenance window, problems were encountered by the web host provider and my web server was offline for a 12-hour period.

The poor performance of the web server and downtime represent an opportunity loss for my website venture/business. New or returning visitors unimpressed with the slow page load or no server response during the 12-hour period following the scheduled maintenance window might decide to skip my website, and not return for another look.

In the telecommunications industry, telephone service in the PSTN (public switched telephone network) delivers quality and performance tuned to five-9s uptime availability. That is 0.99999 uptime which represents about 5.26 minutes of downtime per year. People may not realize or appreciate the uptime availability of plain old telephone service (POTS) which operates even during electrical power outages (the telephone switching equipment has its own independent power supply).

The Internet runs on a networking protocol known as IP (Internet Protocol). The protocol itself operates on a "best-effort" basis which means data is dropped if the network is unable to deliver the information, without any attempt to resend the information. If the network is congested, dropping data packets is the solution to alleviate the congestion. The servers which make up the Internet are themselves not of five-9s reliability. To achieve higher levels of uptime requires added expense for system redundancy (e.g. duplicated hardware to take over in case of a failure).

This recent experience in web server downtime is a reminder that despite the very reliable computing hardware (routers and servers) running the Internet, we are all exposed to system failures. This event was disappointing. I am still working on correcting other things that have stopped working since the system recovery. e.g. getting automatic stock data update to work again.

Blog Entries Listing
2010  Aug  Sep  Oct  Nov  Dec 
2011  Jan  Feb  Mar  Apr  Jun  Sep  Oct  Dec 
2012  Jan  Feb  Apr  May  Jun  Oct  Nov  Dec 
2013  Jan  Feb  Mar  May  Jun  Dec 
2014  Jan  Mar  Jun  Jul  Aug  Oct  Nov  Dec 
2015  Jan  Feb  Mar  Apr  May  Jun  Oct  Nov  Dec 
2016  Jan  Feb  Mar  Apr  May  Jun  Jul  Aug  Sep  Oct  Nov  Dec 
2017  Jan  Feb  Apr  Oct 
2018  Jan  Feb  Mar  Apr 
2020  Mar  Apr  May 

Visit igtsoft.com for
your photography needs.


The only reason for time is so that everything doesn't happen at once. - Einstein