Computing

Tier-2 site:

CIŚ Computing Centre provides 4320 computing cores (35 000 kHS06 / month), 13,5 TiB RAM and more than 750 TB of disk space, for the needs of the experiments conducted at the European Organization for Nuclear Research CERN. It plays a role of the Tier-2D site and is the 7th most important computing centre for the LHCb experiment (provides around 5% of the total normalized CPU time). The main job types are Monte Carlo simulations and user analysis.

Within the framework of European Grid Infrastructure (EGI) CIŚ Computing Centre provides:

  • 108 blade servers (Supermicro-brand), each equipped with:
    • 40 HT computing cores (2x Intel Xeon E5-2680 v2 processors);
    • 128 GiB of RAM (DDR3);
    • 1 local 400 GB SSD;
    • 2 network adapters (10-Gigabit Ethernet);
    • 1InfiniBand QDR ports (40 Gbit/s per port);
    • The theoretical performance of a single node: 448 GFLOPS.
  • Data storage:
    • Based on Netapp FAS6280 and FAS6290 nodes;
    • Equipped with 3TB and 4TB SAS HDD;
    • The total capacity available to the grid users: 750 TB.

Pilot Loggers for Dirac

DIRAC (Distributed Infrastructure with Remote Agent Control) INTERWARE is a software framework for distributed computing providing a complete solution to one (or more) user community requiring access to distributed resources. DIRAC builds a layer between the users and the resources offering a common interface to a number of heterogeneous providers, integrating them in a seamless manner, providing interoperability, at the same time as an optimized, transparent and reliable usage of the resources. The DIRAC platform was adapted as a computing solution by the high-energy physics experiments like LHCb, Belle-2, but also by many European projects including BioMedical, which use it as a platform to perform the advanced GRID operations.

Our group is involved in the development and the implementation of the distributed system of the so-called pilot loggers for the DIRAC platform. The main objective is to provide scalable and distributed system of the logging network. For this purpose we developed the architecture based on the message queue system using the RabbitMQ server.