PROGRAM

Visual analytics for monitoring distributed computing systems for mega-science experiments

Modern experiments in astronomy, particle physics, and astrophysics deal with data on a scale of hundreds of petabytes / exabytes. Working with data of this scale is impossible without the use of a global computing infrastructure. Thus, the distributed computing infrastructure for analysis, processing and management of experimental data at CERN has more than 170 computing centers around the world, in addition to grid resources, it includes cloud computing resources and supercomputers. The total number of servers! interconnected by high-speed communication channels (10 Gb / s), estimated at 130 thousand. Managed data volume is over exabyte. About 5 PB of data are transferred between the centers and are used in physical analysis tasks daily.

To monitor and analyze the work of the computing infrastructure, a multi-level system was created that is constantly evolving, including using machine learning technologies (similar volumes of data and processing methods will be typical for LSST, SKA, and NICA experiments). Our research team works closely with the ATLAS and CMS experiments on the LHC, and is involved in the development of new methods and tools for monitoring and data analysis. As part of this cooperation, a number of promising areas for the development of monitoring tools and data analysis methods were identified, which will be presented in the report.

 Due to the self-isolation regime, the seminar will be held in the form of a webinar on the Zoom platform.

To participate in the webinar, we ask you to enter your data in the Google spreadsheet:

https://docs.google.com/spreadsheets/d/1ffB3TL8szJnIf6NHwtI2NihlWW_2c-XBt741ZZwkNVg/edit?usp=sharing

A link to the Zoom conference will subsequently be sent to your email address.

Instructions for installing and using the Zoom platform are available here:

https://support.zoom.us/hc/ru/articles/201362033-Начало-работы-на-ПК-и-Mac