Niall Kennedy presents dimensions of Google's grid based on an analysis of a research paper by its engineers.
"Google currently processes over 20 petabytes of data per day through an average of 100,000 MapReduce jobs spread across its massive computing clusters. The average MapReduce job ran across approximately 400 machines in September 2007, crunching approximately 11,000 machine years in a single month."
Google is estimated to have spent over almost $ 2.5 billion in Capex in 2007. As I have written before this growing SaaS capacity means moving Capex away from corporate IT to vendors.
In fact, I was surprised when I asked SAP executives in December if they were seeing a spike in capital spend as they prepare for BBD, they said no. But then they are not a SaaS vendor yet. Nothing like Google. Nobody is.
