On-Demand Services

DST offers senior level engineers, architects, and developers to help you rapidly and predictably launch and manage your HPC vision. Our unique teaming model fully engages you in the process and acts as an extension of your team, working with your HPC staff members and the HPC vendor community. DST Services encompassing best practices to deploy, manage, trouble shoot or tune premium products like Bright Cluster Manager, IBM Spectrum Scale, Lustre, BeeGFS, Quobyte, ZFS, DDN, Penguin, Mellanox, Seagate ClustorStor, RAID Inc, Cray and others.

On Demand Services offer our clients the ability to scale out senior level expertise focused on the effort at hand, becoming an extension of your team. Having an agreement in place with Data in Science for a number of hours permits easy consumption access to on-demand service for fulfilling organizational needs. Typical use cases for HPC on-demand services include:

  • Project-based Assistance – Our team of senior HPC engineers work with your team to provide HPC support to accelerate the deployment of targeted efforts and projects including file system migrations, R studio deployment, enabling MatLab, additional cluster builds, or establishing a swift stack archive repository.
  • Internal Skill Augmentation – On-demand hours are often employed to augment the existing staff. Organizations often only have a single member of the team who poses advanced understanding in parallel file system, Linux administration, or cluster management experience. Our ability to augment the team when a senior level skilled resource, takes vacation, leaves for training, or transitions to a new job, helps organizations insure a deeper benches of critical skillsets.
  • Lifeline – On-demand Services could be a lifeline support system for your existing HPC team. Our highly skilled HPC centric team acts as the extra set of eyes in HPC Support to optimize the cluster or troubleshoot a particularly difficult issue with rapid resolution. A resolution that often takes days or weeks just to research.

The following is a non-exhaustive list of skills that the DST team can perform with an On-Demand engagement:

  • File System: GPFS, Lustre, XtreemFS, Quobyte, ZFS
  • R/RStudio/Parallel R
  • LSF, PBS, SGE and UGE Schedulers
  • Bright Computing Cluster Manager
  • Python and Perl Scripting
  • JAVA Development
  • MATLAB and associated MATLAB Modules
  • Linux Engineering
  • Network Engineer
  • KVM, oVirt and OpenStack Virtualization
  • DataDogHQ and Nagios
  • Cluster and File System Architects, Engineers and Administrators

Data in Science’s field proven On-demand methodology provides focused expertise as an extension of your team providing a high level of accountability in HPC Support; including real-time collaboration and detailed weekly reports on all activities for your review.

on-demand-services-graphicOn-Demand Methodology:

  • Contract established with hours to be consumed on demand
  • Project-based work SOW
  • Detailed project plan
  • Collaboration tools and constant communication
  • Weekly reports detailing all activities
  • Documentation of the effort upon completion

The DST team delivers accountability to your organization with a custom solution to act as an extension of your team. The following illustrates a report that is weekly report that is designed to inform the client of the hours used and the detailed tasks performed during those hours.


Who is DST? Our customers say our best feature is we’re flexible, nimble, and fast in making the experience for your researchers the best it can be.