Google, Inc. is developing a powerful and complex data-warehousing system the company dubbed "Mesa" to process a massive amount of data in the shortest time possible.

Google's Internet advertising has led to the development of Mesa. To serve its internal needs and advertising customers, Google gathers ad data then records and processes the detailed information in real time.

Like Us on Facebook

Mesa takes in the information at a rate just below real time and processes the data on a great scale, according to Venture Beat.

"Mesa handles petabytes of data, processes millions of row updates per second, and serves billions of queries that fetch trillions of rows per day," according to the authors of the paper.

A petabyte's worth of information is equivalent to 1,000 terabytes or 1 million gigabytes.

They also added that Mesa is immune to datacenter-failures because it is "geo-replicated across multiple datacenters."

Mesa might also advance into a new cloud service accessible through the Google Cloud Platform.

The system could aid Google distance itself from competition such as Amazon Web Services, which has a similar service to Mesa dubbed "Redshift," and Microsoft Azure which is capable of dropping cloud service prices.

Azure can readily decrease the cost of its service as it often releases new cloud services, just like Google.

Mesa also separates itself from the query engine Presto, which the social networking site Facebook developed to cope with latency issues that Hive could not deal with.

The Mesa system might be particularly well suited for deployment in data centers around the globe.

"The cloud computing paradigm in conjunction with a decentralized architecture has proven to be very useful to scale with growth in data and query load," they wrote. 

TagsCloud computing services, Cloud computing, Amazon, Amazon Web Services, Azure, Microsoft Azure, Research