Enabling automated HPC / database deployment via the appscale hybrid cloud platform

  • Authors:
  • Chris Bunch;Chandra Krintz

  • Affiliations:
  • University of California, Santa Barbara, Santa Barbara, CA, USA;University of California, Santa Barbara, Santa Barbara, CA, USA

  • Venue:
  • Proceedings of the first annual workshop on High performance computing meets databases
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we discuss a prevalent issue facing the HPC community today: the lack of automation in the installation, deployment, and integration of HPC and database software. As a result, scientists today must play a dual role as researchers and as system administrators. The time required for scientists to become proficient with software stacks is significant and has increased with the complexity of modern systems such as cloud-based platforms and infrastructures. However, cloud computing offers many potential benefits to HPC software developers. It facilitates dynamic acquisition of computing and storage resources and access to scalable services. Moreover, cloud platforms such as AppScale abstract away the underlying system and automate deployment and control of supported software and services. As part of this project, we have extended AppScale with domain specific language support called Neptune that gives developers straightforward control over automatic configuration and deployment of cloud applications. Neptune also extends cloud support beyond web-services to HPC applications, components, and libraries. We discuss AppScale and Neptune, and how they can be extended via more intelligent database usage to provide a better solution for the next-generation of cloud-based HPC and data-intensive applications.