Oh, so when working at Google, you have to book time on the mainframe (cloud)? Just like the old days with punch-cards...
Google and North Caroline State University researchers have worked out how to instrument cloud infrastructure to the point where they can predict future demand 68% per cent better than previously, giving other cloud providers a primer for how to get the most out of their IT gear. The system was outlined in an academic paper …
The AGILE paper describes a research project that was done at North Carolina State University, not at Google. My NCSU colleagues deserve all the credit for developing the algorithm and evaluating it; I only assisted with writing the paper.
The AGILE software has never been run at Google. The research used publicly-available trace data from a Google compute cluster (see http://googleresearch.blogspot.com/2011/11/more-google-cluster-data.html for more information, or search for [more google cluster data]). There is no connection between AGILE and the CPI^2 infrastructure.
FYI, the definitive version of the paper appears in the ICAC proceedings: https://www.usenix.org/conference/icac13/agile-elastic-distributed-resource-scaling-infrastructure-service
Biting the hand that feeds IT © 1998–2021