Recruiting additional resource for hpc simulation

a technology of simulation and additional resources, applied in the field of computing clusters and cloud computing, can solve the problems of bored people walking away, most people and organizations can not afford to purchase and maintain computing systems,

Inactive Publication Date: 2012-05-17
CREATIVEC
View PDF7 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

There are problems with the statically defined virtual machines such as hardware failures and changing the virtual machine by adding or removing computers.
Too slow a simulation means the person gets bored and walks away.
Most people and organizations can not afford to purchase and maintain computing systems having enough power that a person can interact with an MD simulation and receive near real-time feedback.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Examples

Experimental program
Comparison scheme
Effect test

first embodiment

[0037]In a first embodiment, the user's computer can be a very powerful machine having numerous GPU units. This is the simplest case where all the applications run on that one machine.

second embodiment

[0038]In a second embodiment, the user also has a GPGPU cluster. NAMD (or LAMMPS) is distributed throughout the cluster and streams data to the user's computer running VMD. This may require a fast link between cluster and display computer. This is particularly true when the displaying the animation in 3D. A slower link can be used if the cluster runs VMD and uses VirtualGL to display to the user.

third embodiment

[0039]In a third embodiment, a remote cluster can run NAMD, VMD, and VirtualGL while the user's computer runs only VirtualGL. A reasonably fast and low lag internet connection can provide adequate interactive MD simulation and visualization.

[0040]In a fourth embodiment, the user begins with the second embodiment detailed above but wants more speed. The user can recruit a remote GPGPU cluster. Recruitment itself is discussed above. Portions of the MD simulation can be simply passed over to the remote cluster. One way to obtain this is to momentarily stop the simulation, reconfigure it to use all the resources, and restart it. This may require a fast low lag connection to the remote cluster. Another way is to stop the simulation and send it completely to the remote cluster such that the user, in essence, switches from using the second embodiment to using the third embodiment. Yet another way is for the simulation to examine the multiprocessing architecture (eg the MPI configuration) e...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

Graphics processing units (GPUs) deployed in general purpose GPU (GPGPU) units are combined into a GPGPU cluster. Access to the remote GPGPU cluster is then offered as a service to users who can use their own computers to communicate with the GPGPU cluster. The users' computers can be standalone desktop systems, laptops, or even another GPGPU cluster. The user can run a parallelized application locally and patiently wait for results or can dynamically recruit the remote GPGPU cluster to obtain those results more quickly. Dynamic recruitment means that the users can add remote GPGPU resources to a running application.

Description

CROSS REFERENCE TO RELATED APPLICATIONS[0001]This patent application is a continuation in part of U.S. patent application Ser. No. 12 / 895,554, entitled “GPGPU SYSTEMS AND SERVICES” which was filed on Sep. 30, 2010 and which is incorporated herein by reference in its entirety.TECHNICAL FIELD[0002]Embodiments relate to computing clusters, cloud computing, and general purpose computing based on graphic processor units. Embodiments also relate to massive computing power offered on a subscription basis. Embodiments additionally relate to profiling massively parallel programs on a variety of cluster configurations.BACKGROUND OF THE INVENTION[0003]Parallelization is a well-known way to more rapidly obtain computational results. The reason is quite simply that two processors or computers working concurrently on a job should finish more quickly than if only one of the computers is used. One of the currently popular techniques for obtaining parallelization is to connect a great many computers...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(United States)
IPC IPC(8): G06T11/20G06F15/16
CPCG06F9/5011G06F2209/508G06F2209/504Y02B60/142Y02D10/00G06F2209/505G06T1/20G06F30/20
Inventor SCANTLEN, GREGSCANTLEN, GARY
Owner CREATIVEC
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap