Technologies/Resources

Servers

  • High-Performance Computing
    High-performance computing (HPC) resources are available for no charge to Cedars-Sinai staff. The primary HPC system utilizes the Son of Grid Engine job scheduler, which is derived from Sun Grid Engine, developed by Sun Microsystems.

    This is a cluster of IBM-Intel-based computers that utilize Son of Grid Engine as the job scheduling and management service. 

    This system is best suited for tasks that lend themselves to parallel processing. The overall configuration consists of 342 CPUs and 1.38 TB of memory. The basic compute node has 3 Ghz CPUs and 16 GB of memory, while some “high memory” nodes consist of a mix of 2.4-3.5 Ghz CPUs and between 64-80 GB memory. Additional nodes are added to the cluster on a routine basis.

    Applications installed on the cluster include:
    • FSL and Freesurfer (FMRI, MRI, and DTI brain imaging)
    • NAMD (simulation and modeling of molecular systems)
    • MATLab / Octave (scientific programming and visualization tool)
    • R (statistical computing and graphics)
    • JAGS (Bayesian model analysis, Markov Chain Monte Carlo simulation)
    • Bioperl (Perl modules for bioinformatics computation)
    • MACH (resolve long haplotypes)
    • NCBI BLAST (sequencing analysis)
    • Clustal W (sequence alignment)
    • Fasta (search protein and DNA sequence data)
    • GROMACS (molecular simulation)
    • breakdancer (genomewide detection of variants)
    • FFW3 (discrete fourier transform)
    • QIME (comparison and analysis of microbial communities)
    • TMAP (programs to build genetic maps)
    • EMBOSS (tools used for sequence analysis)
    • Fortran (compiler)

Other software can be added as requested subject to compatibility requirements as determined at the time of the request.

    • VMware Cluster (internal)
      This is a cluster of IBM 3640 servers with 80 CPUs (3.3 Ghz) and over 1.6 TB of memory that can be used to form VMWare virtual servers, either Linux or Windows. The cluster sits behind the firewall and is faster to install than a physical machine. There is no charge for a virtual machine, but there may be some architectural limits on the number of CPUs and/or memory that can be allocated.
    • VMware Cluster (external)
      This is a cluster of IBM 3640 servers with 36 CPUs (3.3 Ghz) and 296 GB memory that can be used to form VMWare virtual servers, either Linux or Windows. This cluster is intended for use with external-facing (i.e. outside the firewall) applications.


      All VM requests are subject to an EIS review to ensure compliance with technical and security standards.

      Request access (for internal users only)


    Disk Storage

    • Network Attached Storage
      All investigators are provided access to centrally managed disk storage that is backed up daily. More than 700 TB of storage is available, configured as network attached storage.
    • High-Performance Storage
      High-performance storage is provided by Isilon X200 disk systems. This storage is intended for use primarily with the HPC cluster, although it is also available for large data sets that exceed the 5 TB architectural limit. Those data are automatically replicated at a secondary data center in Phoenix. Please email RISCC staff for more information.
    • SharePoint
      Another option for file sharing is SharePoint. This Microsoft application is optimized for sharing Word documents, Excel spreadsheets and similar.

    Request access (for internal users only)


    Collaboration

    Options for sharing files with investigators at other institutions include:

    • Collaboration Server
      A special collaboration server is available for sending large files (up to 20 GB). For large files, Internet bandwidth may be a limiting factor. Typically, it takes about 15 minutes to upload a 5 GB file to the collaboration server from an on-campus workstation; the download speeds at the recipient end will vary depending on the type of Internet service at the remote end. All files sent via the collaboration server are automatically encrypted both "in transit" and "at rest," so it is safe to use this for PHI and other confidential information.
    • Box.com
      Cedars-Sinai has a HIPAA business associate agreement in place with Box.com; all data sent to Box.com is encrypted "in transit" and "at rest," so it is safe to use Box.com to share data.
    • Other "Cloud Providers"
      Other options such as Google and Amazon Web Services may be available, depending on the requirements. Please contact the RISCC for details.

    Request access (For internal users only)


    Application Development Platforms

    • REDCap
      REDCap is available primarily as a self-service research database development tool. Online tutorials are available. REDCap is best suited for research projects where a data entry form is accessed via a web browser and/or where follow-up reminders and such are required. When used in self-service mode, REDCap enables researchers to quickly configure a database by uploading database fields via a spreadsheet or with a wizard that guides the user through the process. Once the database is ready, the user can contact RISCC to have the database put into production.

    Quick Start Guide (for internal users only)

    • TRICS
      TRICS is an EIS-developed legacy database platform that is available to researchers who require an internal research repository. TRICS requires an EIS analyst to setup and can interface with other hospital clinical systems by way of batch imports.This is not a real-time import but a "snapshot in time" import.

    Request access (For internal users only)


    Clinical Research Tools

    • Los Angeles Data Resource
      The Los Angeles Data Repository is an application used for cohort identification among a consortium of institutions (currently Cedars-Sinai, UCLA and the University of Southern California). The application is based on i2b2 and SHRINE and generates criteria specific de-identified data and counts of cohorts at each institution.

    Request access (for internal users only)