Collaborative Editing Through MXF Server System
Computer scientists have developed a new way to optimize storage systems that dramatically improves processing efficiency and output speed.
Modern storage systems use only one server, which slows down the retrieval of information displayed to the user. In this case, the backup server remains inactive until the main.
A team of programmers from the University of Waterloo decided to use a new approach to optimize storage, called FLAIR, that uses all available servers on the network. As a result, when the user makes a request, if the main server is full, then the other is automatically activated to fulfill it..
Implementation of programmable networks is a key factor in supporting FLAIR, scientists say. The protocol developed uses this technology to create an intelligent network layer that can find the fastest way to fulfill queries for information retrieval. The team’s assessment showed that the new method allows execution of queries 2.5 times faster than classic projects..
After formally verifying that FLAIR was functioning correctly, the team tested it under real-world workload conditions on a university campus with a programmable network cluster. New protocol increased search speed by 35-97%.
Scientists argue that FLAIR is able to significantly improve the performance of databases and information processing engines, which are the backends for banking and medical systems, as well as financial transactions. The new approach can also be applied to any modern cloud application..
We also previously reported that at MIPT developed a new system of high-speed data transmission over long distances.
text: Ilya Bauer, photo: Unsplash