• Set ALLOWTOPICCHANGE = TWikiAdminGroup,Main.LCGAdminGroup
• Set ALLOWTOPICRENAME = TWikiAdminGroup,Main.LCGAdminGroup #uncomment this if you want the page only be viewable by the internal people #* Set ALLOWTOPICVIEW = TWikiAdminGroup,Main.LCGAdminGroup

Specifications, standard Benchmark values

A summary of the resources CSCS is pledged to deliver in April 2021 as agreed with the CHIPP Computing Board:

• They are calculated using projections from the RRB (resource review board) and agreed with SNF.
• Above resources differ from CRIC due to a misunderstanding. This was clarified with the WLCG office in November 2020.
• The extension of the foreseen storage capacity was not available until July'21 due to problems in the chip supply chain.

CPU

HS06 = HEP-SPEC06 benchmark value (see details in HEP-SPEC06 webpage) (Note: 1 HS06 = 1 GFlop, approximately)

Pledge compute capacity is 10% lower due to various inefficiencies.

These CPU resources are divided among the VOs with fairshare (data taken from the last week of usage) and a few reserved slots (for exclusive usage):

• ATLAS: 40% fairshare (max of 2000 jobs running, 4000 in queue).
• CMS: 40% fairshare (max of 2000 jobs running, 4000 in queue).
• LHCb: 20% fairshare (max of 1500 jobs running, 4000 in queue).

Last update: August 2018

Storage infrastructure

Central Storage - dCache

The dCache services are provided by 12 servers (10 "Storage Elements" and 2 "Head Nodes")
Storage Elements are connected to the storage systems through the CSCS SAN

Total space available on dCache is 4.6 PB and is distributed in 40% - 40% - 20% (ATLAS/CMS/LHCb) fashion.

Last update: November 2019

Scratch - Spectrum Scale

The 'scratch' filesystem is available to Compute nodes through special DVS nodes and is provided by a Spectrum Scale cluster of 16 nodes (4 meta + 12 data).

Metedata is located on Flash Storage

The filesystem is featuring a policy based pool tiering
Tier1: SSD based data pool (90TB) where all files are written
Tier2: HDD based data pool (380TB) where less frequently used data is moved (this mechanism is transparent for the clients)

Last update: November 2019

Network connectivity

External network

CSCS has a redundant 100Gbit/s connection (provided by SWITCH). LHConCray computing resources, storage and all related services are connected to the Internet with many high speed interfaces.

Cluster Management Nodes

Most of the service nodes are virtualized. To see a complete list please visit the FabricInventory#Virtual_Machines

Individual services are described in ServiceInformation