All news

Russian scientists to deliver data base system for Large Hadron Collider

At the moment CERN physicists have to use relational data bases - ORACLE and MySQL while social networks have developed their own technologies

TOMSK, September 2. /TASS/. Researchers at the Tomsk Polytechnical University are working on a data storage system that will allow for handling tremendous amounts of data gathered in experiments being staged at the Large Hadron Collider of the European Organization for Nuclear Research (CERN), the head of the physical software group at the United States’ Brookhaven National Laboratory cooperating with CERN, Aleksey Klimentov, told the media.

"At the moment CERN physicists have to use relational data bases - ORACLE and MySQL, while social networks, such as Facebook, and the Google and Yandex search engines have developed their own technologies, which they are prepared to provide on a commercial basis. But then it will be an enigmatic ‘black box.’ Google, for instance, prefers to keep its operation closed for outsiders. That’s classified information. In the meantime, the scientists using the Large Hadrone Collider should be certain about the confidentiality of their data," Klimentov said.

At the moment the experiment called ATLAS which Klimentov cooperates with requires managing 160 petabytes of data. None of the existing approaches promise a suitable solution to the problem. CERN and Tomsk Polytechnic University scientists have created in Tomsk what they have called the Big Data Analytics and Technologies Laboratory, charged with the task of delivering within a two-year deadline a new solution of storing large amounts of data.

The data base will employ both existing principles of storing large amounts of data.

"Table storage constitutes the classical approach. But such a data base cannot be scaled. It is created for a certain amount of data, which is no longer relevant in many spheres. The other, scalable data storage, was developed for the social networks. Social networks keep data as separate files," Klimentov said.

It is expected that the system yet to be developed in Tomsk will be tested on the megadata from one of CERN experiments. If successful, it will be offered to industrial companies handling large amounts of data.