Whatever else you might say about Google, they understand how to engineer web-scale systems. Check out Greg’s notes (or Michael Bendersky’s via Jeff’s Search Engine Caffe) about Google Fellow Jeff Dean’s keynote at last week’s WSDM 2009 conference.
Here’s the teaser from Greg’s notes:
- Advertisement -
Jeff gave several examples of how Google has grown from 1999 to 2009. They have x1000 the number of queries now. They have x1000 the processing power (# machines * speed of the machines). They went from query latency normally under 1000ms to normally under 200ms. And, they dropped the update latency by a factor of x10000, going from months to detect a changed web page and update their search results to just minutes.
This is no small feat, and it gives you a sense for the bar that Google has set in the web search market. Students of the Innovator’s Dilemma take note: if you want to beat Google, you’re not going to do it by stuggling to incrementally outdo them on their own turf. For the use cases it addresses, Google is surely good enough. And damn fast.
Whatever else you might say about Google, they understand how to engineer web-scale systems. Check out Greg’s notes (or Michael Bendersky’s via Jeff’s Search Engine Caffe) about Google Fellow Jeff Dean’s keynote at last week’s WSDM 2009 conference.
- Advertisement -
Here’s the teaser from Greg’s notes:
Jeff gave several examples of how Google has grown from 1999 to 2009. They have x1000 the number of queries now. They have x1000 the processing power (# machines * speed of the machines). They went from query latency normally under 1000ms to normally under 200ms. And, they dropped the update latency by a factor of x10000, going from months to detect a changed web page and update their search results to just minutes.
This is no small feat, and it gives you a sense for the bar that Google has set in the web search market. Students of the Innovator’s Dilemma take note: if you want to beat Google, you’re not going to do it by stuggling to incrementally outdo them on their own turf. For the use cases it addresses, Google is surely good enough. And damn fast.