NSF CC* Data Storage

High Volume Data Storage Infrastructure for Scientific Research and Education at ʳɫÊÓƵ State University Shared as Open Science Data Federation Data Origin

The  program invests in coordinated campus-level cyberinfrastructure improvements, innovation, integration, and engineering for science applications and distributed research projects.

The ʳɫÊÓƵ proposal intends to develop a high-capacity storage system with around 4.3PB overall usable storage as an Open Science Data Federation (OSDF) data origin to support scientific research and education activities on both campuses of ʳɫÊÓƵ (ʳɫÊÓƵ). 30% of the storage is to be allocated for hosting datasets of external researchers and sharing ʳɫÊÓƵ spawned datasets to empower national research projects.  You can learn more in the Ê³É«ÊÓƵ's Proposal Abstract and in the .

Updates

October 2024: need confirmation from UCSD about OSG connectivity​

September 2024: Routing resolved with SOX and Cisco support.​

August 2024: Troubleshooting issue with dropped frames to UCSD.​

July 2024: Kubernetes configuration for OSG sharing as a Origin Node. ​

June 2024: SOX will switch to jumbo frames for Internet2 pipeline.

June 2024: Transceivers for 25Gbps connectivity are expected to be received and tested.

May 2024: SDSC has been given access for Kubernetes node configuration.

May 2024: CC* K8 node connected to 100 Gbps connection for setup and configuration (to be downgraded to 25 Gbps for production).

May 2024: Inter-campus link upgraded to 100Gbps

April 2024: Campus link to Internet2 upgraded to 100Gbps

April 2024: Seminar - ACCESSing Advanced National Supercomputing and Storage Resources for Computational Research: description, slides and video available.

March 2024: Investigating network switch issues with a variety of transceivers.

March 2024: Open Data Committee Meeting: Planning for upcoming seminars.

February 2024: Resolving issues, cluster is reporting as healthy

February 2024: Issue with proxy and a disk reporting as bad.

February 2024: Configurations for CRUSH maps and erasure encoding.

February 2024: Proxies created for internal and external networks

January 2024: cephadm Orchestration tool configured for OSD daemons

January 2024: Disk discussion with UCSD and OSDs manually configured to use NVME disks

January 2024: Ceph installation started

January 2024: Benchmarking: network and I/O

December 2023: Docker installation/configuration

December 2023: System configuration for private network

December 2023: OS installations

November 2023: Privatvate network created for system

October 2023: Hardware installed in server room

September 2023: Hardware arrived in ʳɫÊÓƵ Receiving

September 2023: NSF CC* Workshop

September 2023: OpenData Committee formed

September 2023: Introductions to UCSD contacts

August 2023: ʳɫÊÓƵ hardware purchase completed

Summer 2023: Internet2 connection to campus made