Research and Scientific Computing at USC maintains redundant arrays of networked storage used for housing data safely and with minimal downtime.
There are two categories of data repositary available within RCC storage system: Home on HPC clusters, data shares mounted on SAN storage.
Home on HPC clusters
Each HPC cluster user has a home directory. It is accessible from all computer nodes in Planck and ACM-CHEM. User guideline are listed below.
This directory is generally used for storing computational research projects, source data and results.
Data share on SAN storage
To support the file repository needs for researchers, RCC has deployed a SAN file archival system. The SAN system is a complex disk-based storage system with capacity of 120 Terabytes and minimal downtime.
Storage - P4500 G2 120TB
Storage Clustering for disruption free growth
Network RAID 0, 5, 6, 10, 10+1, and 10+2 (Synchronous Replication)
Application Integrated Snapshots
Remote Copy (Asynchronous Replication with Bandwidth Throttling)
Multi-Site HA/DR Solution (Synchronous Replication across several locations)
Online Volume Migration and automated Software Upgrades
Self-Healing Storage: Proactive disk / volume scrubbing with automatic bad block repair
Network RAID Levels (per volume) 0, 5, 6, 10, 10+1, and 10+2 protect against double disk, controller, storage node, power, network, or site failure
Hardware RAID Levels 10, 5, 6 (default is level 5)
Hot-plug hard drives
Redundant power supplies
Integrated storage controller w/ battery-backed cache
Hyper-redundant clustered storage (achieved in combination with the P4000 SAN/iQ storage software)
Redundant stackable HP networking switches for iSCSI data w/ 10GbE capability
Applying for a storage allocation
Any faculty and researcher in CAS can apply for a storage allocation. For precise information on policies, please send an email to firstname.lastname@example.org. RCC Staff propose multiple level of backup schema to ensure safety of the research data.
SAN user guide