+7 (812) 622 16 80

RAIDIX Expert Received Bertrand Meyer Award at SECR 2017

31.10.2017

Head of RAIDIX Research Lab Svetlana Lazareva won a Bertrand Meyer Award at the SECR 2017 international conference (St Petersburg, Russia, October 20–21, 2017). As a speaker, Svetlana delivered a presentation titled “Smart face control: machine learning algorithms for efficient SSD caching”. Data storage vendor RAIDIX has participated in SECR since 2014. In 2015, RAIDIX’s paper also succeeded in securing the first award at the conference.

SECR is a major Russia-based IT event shaped up by the international program committee – industry experts, scholars and researchers from the US, European Union, and the CIS. SECR is a unique networking event featuring a host of hot topics, cutting-edge research results, and cross-specialty knowledge transfer. All submitted papers undergo a meticulous selection process, only 35 percent of suggested presentations are granted acceptance by the committee.

Svetlana Lazareva’s “Smart face control” described a new algorithm of SSD cache filling. The algorithm builds on analyzing incoming requests to the data storage system with the use of machine learning. The goal of this research is to extend the lifecycle of solid-state drives when the latter are utilized as caching devices.

Experts at the RAIDIX Research Lab are authors to multiple technology patents registered in the US. The patented RAIDIX algorithms include new erasure coding methods, parity RAIDs 6 and 7.3, advanced (preemptive) data reconstruction, the intellectual QoSmic module for balancing performance on the application level, and more works.

The Lab engages in development of new algorithms and features for the company’s flagship product RAIDIX – management software for data storage servers, and makes inroads into distributed storage techniques, deep learning, data mining, software-defined memory, and other promising technologies. Brand-new research results form part of the RAIDIX functionality ensuring greater performance and fault-tolerance in versatile data-intensive industries.