<body><script type="text/javascript"> function setAttributeOnload(object, attribute, val) { if(window.addEventListener) { window.addEventListener('load', function(){ object[attribute] = val; }, false); } else { window.attachEvent('onload', function(){ object[attribute] = val; }); } } </script> <div id="navbar-iframe-container"></div> <script type="text/javascript" src="https://apis.google.com/js/plusone.js"></script> <script type="text/javascript"> gapi.load("gapi.iframes:gapi.iframes.style.bubble", function() { if (gapi.iframes && gapi.iframes.getContext) { gapi.iframes.getContext().openChild({ url: 'https://www.blogger.com/navbar.g?targetBlogID\x3d8211560\x26blogName\x3dTech+Tips,+Tricks+%26+Trivia\x26publishMode\x3dPUBLISH_MODE_BLOGSPOT\x26navbarType\x3dBLUE\x26layoutType\x3dCLASSIC\x26searchRoot\x3dhttp://mvark.blogspot.com/search\x26blogLocale\x3den\x26v\x3d2\x26homepageUrl\x3dhttp://mvark.blogspot.com/\x26vt\x3d-5147029996388199615', where: document.getElementById("navbar-iframe-container"), id: "navbar-iframe" }); } }); </script>

Tech Tips, Tricks & Trivia

by 'Anil' Radhakrishna
An architect's notes, experiments, discoveries and annotated bookmarks.

Search from over a hundred HOW TO articles, Tips and Tricks


Google Search - Then & Now

Speaking at WSDM 2009, Google Fellow Jeff Dean in a keynote talk shared how Google has grown from 1999 to 2009. Various search engine enthusiasts have blogged about the facts presented in the talk from their own notes.

Excerpts:
  • Google handles 1000 times more queries now with 1000 times more processing power (# machines * speed of the machines).
  • It uses 1000 machines to handle a single query compared to just 12 previously.
  • Query latency has improved from under 1000ms to normally under 200ms now.
  • Crawler updates now take minutes compared to months in 1999.
  • For many pages, search results now change within minutes of the page changing.
  • Rolled out seven major rearchitecture efforts in ten years.
  • In-house design from the ground up: rack design, pc class motherboards, linux, and in-house software (GFS, BigTable, etc...)
  • Google's machine translation models use a million lookups in a multi-terabyte model just to translate one sentence.
Popularity comes at a price. Google is now accused of harming the environment with it's energy consumption (a typical search uses "half the energy as boiling a kettle of water" and produces 7 grams of CO2) and for not practicing what it preaches.

Also see: How LARGE websites manage performance & scalability

Labels:

Tweet this | Google+ it | Share on FB

« Home | Next »
| Next »
| Next »
| Next »
| Next »
| Next »
| Next »
| Next »
| Next »
| Next »

»

Post a Comment