System Types

From Cheaha
Revision as of 21:11, 29 November 2007 by Jpr@uab.edu (talk | contribs) (Create high-level summary of UABgrid node types)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search


Attention: Research Computing Documentation has Moved
https://docs.rc.uab.edu/


Please use the new documentation url https://docs.rc.uab.edu/ for all Research Computing documentation needs.


As a result of this move, we have deprecated use of this wiki for documentation. We are providing read-only access to the content to facilitate migration of bookmarks and to serve as an historical record. All content updates should be made at the new documentation site. The original wiki will not receive further updates.

Thank you,

The Research Computing Team

UABgrid can be visualized as a computing platform composed of several types of systems. These systems implement a variety of services and together provide integrated environment for building research workflows. These types are conceptual and don't require strict separation, that is, one computer could take on the role of several types at once.

  • Computation Node - this is one of the primary system types and is key to carrying out computationally intense jobs. The HPC clusters on campus are prime examples of this sytem type. While users will typically access a computation node in one of the primary compute facilities on campus, it is also possible to integrate a local, smaller scale compute farm into this framework so you can more easily migrate your workflow from local dedicated resources to shared computation facilities.
  • Data Node - this system performs the function of data storage for information related to your research workflow. UABgrid currently provides a central data node that can be used to manage the data distribution for research workflows during computation. Ideally, research groups have dedicated data nodes to house their data locally and directly stage appropriate files to computation nodes when workflows are executed.
  • Application Node - application nodes typically host a workflow specific set of applications that orchestrate the workflow for a specific domain through simplified user interfaces, eg. a web based application that accepts input parameters and carries out specific computational tasks. These are the primary targeted interfaces for users on UABgrid.
  • Development Node - this node is intended for people interested in contributing to the development of UABgrid applications or system services. If you are interested in building an application to host on an application node or working on migration for workflows leveraging computation or data nodes, this is a good choice. See the Development Node setup instructions for information on how to create your own instance.
  • Client Node - this is the general node type for user interaction with UABgrid. At a minimum this node is assumed to support a standards compliant web browser running on any platform. It is typically your desktop or laptop computer running Linux, Mac or Windows. The client node runs the typical desktop applications that are part of your research workflow. No special integration with UABgrid is necessary, however, advanced client application can be developed the directly integrate with UABgrid and provide a rich user experience for start to finish workflow integration.