Implementing a large-volume, scaleable backup system in a scientific research environment

ABSTRACT

The Beckman Center at Stanford University Medical Center houses five floors of research laboratories and approximately 80 research groups in Biochemistry, Developmental Biology, Cellular/Molecular Biology and other research areas. In past years a central 2TB disk array system connected to a Solaris Unix server and a separate multi-terabyte Retrospect Server for backing up desktop machines has worked well. However, the explosion of research data, fueled by DNA microarray analysis, realtime PCR, and confocal microscopy has created a demand for a tremendous amount of storage. Specific aspects of the research environment such the high turnover rate of graduate students and postdoctoral students and the demand that research groups be able to purchase computers at their discretion and according to budgetary opportunity set further parameters on the backup system.

The system put in place is a BlueArc Titan 2200 with 9.75 TB of useable drive space. The array is expandable to 512 TB. The disk space on the array can be accessed directly by NFS or Samba mounting. Mac users can also access their data through AppleShare through a Xinet server running on our Sun server.

This presentation reviews the parameters that defined the backup system and outlines challenges that we encountered in the Beckman Center and the Center for Molecular and Genetic Medicine in implementation, user authentication, and user education in the process of deployment.

INDEX TERMS

Primary Classification:
  E. Data
  E.5 FILES
      Subjects: Backup/recovery

Additional Classification:
  C. Computer Systems Organization
  C.4 PERFORMANCE OF SYSTEMS
      Subjects: Reliability, availability, and serviceability; Design studies

General Terms:
Design, Reliability