Speaker
Nikolay Kutovskiy
(JINR)
Description
High demand for the JINR cloud resources facilitated its sufficient growth. That triggered changes needed to be done to overcome problems encountered and to keep QoS for users: main part of computational resources was re-organized as pre-deployed worker nodes of HTCondor-based computing element to decrease the load on OpenNebula services during mass jobs submission, new SSD-based ceph pool for RBD VMs disks with strong disk I/O requirements, dedicated ceph-based storage for the NOvA experiment, re-organized from scratch Infrastructure-as-a-Code approach based on role and profile model implemented with help of foreman and puppet, migration to prometheus-based software stack for resources monitoring and accounting and some other changes.
Primary author
Nikolay Kutovskiy
(JINR)
