
4.5.3 Red Hat Enterprise MRG Grid
MRG
vi
Grid provides high throughput and high performance computing. Additionally, it
enables
enterprises to move to a utility model of computing to help them achieve both higher
peak computing capacity and higher IT usage by leveraging their existing infrastructure to
build high performance grids.
Based on the Condor project, MRG Grid provides the most advanced and scalable platform
for high throughput and high performance computing with capabilities such as:
• Scalability to run the largest grids in the world
• Advanced features for handling priorities, workflows, concurrency limits, usage, low
latency scheduling, and more
• Support for a wide variety of tasks, ranging from sub-second calculations to long
running, highly parallel (MPI) jobs
• The ability to schedule to all available computing resources, including local grids,
remote grids, virtual machines, idle desktop workstations, and dynamically provisioned
cloud infrastructure
MRG Grid also enables enterprises to move to a utility model of computing, where they can:
• Schedule a variety of applications across a heterogeneous pool of available resources
• Automatically handle seasonal workloads with high efficiency, usage, and flexibility
• Dynamically allocate, provision, or acquire additional computing resources for
additional applications and loads
• Execute across a diverse set of environments, ranging from virtual machines to bare
metal hardware to cloud-based infrastructure
www.redhat.com 20
Comentários a estes Manuais