How Google processes 20 petabytes of data each day

January 9th, 2008

The latest CACM has an article by Google fellows Jeffrey Dean and Sanjay Ghemawat with interesting details on Google’s text processing engines. Niall Kennedy summarized it this way on his blog post, Google processes over 20 petabytes of data per day.

“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.”

If big numbers numb your mind, 20 petabytes is 20,000,000,000,000,000 bytes (or 22,517,998,136,852,480 for the obsessive-compulsives among us) — enough data to fill up over five million 4G ipods a day, which, if laid end to end would …

Kevin Burton has a copy of the paper on his blog.

Jeffrey Dean and Sanjay Ghemawat, MapReduce: simplified data processing on large clusters, Communications of the ACM, pp 107-113, 51:1, January 2008.

MapReduce is a programming model and an associated implementation for processing and generating large datasets that is amenable to a broad variety of real-world tasks. Users specify the computation in terms of a map and a reduce function, and the underlying runtime system automatically parallelizes the computation across large-scale clusters of machines, handles machine failures, and schedules inter-machine communication to make efficient use of the network and disks. Programmers find the system easy to use: more than ten thousand distinct MapReduce programs have been implemented internally at Google over the past four years, and an average of one hundred thousand MapReduce jobs are executed on Google’s clusters every day, processing a total of more than twenty petabytes of data per day.

Dean and Ghemawat conclude their paper by summarizing the key reasons why MapReduce has worked so well for Google.

“First, the model is easy to use, even for programmers without experience with parallel and distributed systems, since it hides the details of parallelization, fault tolerance, locality optimization, and load balancing. Second, a large variety of problems are easily expressible as MapReduce computations. For example, MapReduce is used for the generation of data for Google’s production Web search service, for sorting, data mining, machine learning, and many other systems. Third, we have developed an implementation of MapReduce that scales to large clusters of machines comprising thousands of machines. The implementation makes efficient use of these machine resources and therefore is suitable for use on many of the large computational problems encountered at Google.”