May 14th, 2009
Do our Operating Systems need Search Failover?
Outages of search providers can create huge amounts of havoc. But can we build functionality into our OSes to mitigate the problems?
As I write this, Google has failwhaled. The cause of the problem was initally presumed to be at the AT&T Tier 1 ISP level, causing huge amounts of packet loss. (EDIT: This ended up being explained as a routing problem by Google) The net effect of this is that it’s resulted in a crappy day for a lot of us that depend on GMail and use their Google’s search services for just about, well, everything.
Click on the “Read the rest of this entry” link below for more.
When Google goes down, lots of stuff breaks. Not just Google’s own apps like GMail and Google Talk, but also applications like Firefox, which use Google as it’s default search provider.
Default search providers are easy enough to fix in browsers such as Firefox or Internet Explorer (Which uses Windows Live Search) but it requires manual user intervention. However, other applications use Google and its search services. In the event of an apocalyptic business continuity event for a major search provider like Google, how would we as Internet users get all our search services on all our apps working on an alternate provider quickly?
It seems to me that we need some sort of “Search Failover” capability in our desktop OSes and mobile devices — such as a master configuration file which a number of search and Internet service API providers can be plugged into that monitors the health of these providers and would automatically adjust settings globally across all applications that use them in the event of a major outage event. For example, if Google goes down, the configuration would automatically switch to Yahoo or MSN Live Search, depending on the user’s desired configuration. It would also notify the user when the outage occurs and switch back to the original default search provider when the outage is over.
Obviously, Search Failover would not mitigate major systemic outages such as a Tier 1 ISP issue that would affect multiple search services that use it as a bandwidth provider, but presumably all major search providers are procuring connectivity from more than one Tier 1 ISP.
Do we need Search Provider Failover capabilities in our OSes and devices? Talk Back and Let Me Know.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=c49ef12f-c9a6-41e1-8570-60a07403e2fa)
Jason Perlow is a technologist with over two decades of experience integrating large heterogeneous multi-vendor computing environments in Fortune 500 companies. See his full profile and disclosure of his industry affiliations.
Subscribe to Tech Broiler via Email alerts or RSS.
