| 產品介紹 > 中介軟體
> 叢集/網格發展套件> ClusterResources |
| Process More Work in Less Time to Maximize ROI |
．Achieve higher, more consistent utilization of
resource capabilities and capacity with
intelligent scheduling that matches job requests
to best-suited resources
．Use optimized data staging to ensure remote
data availability is synchronized with resource
availability to minimize resource blocking
．Achieve better performance with automatic
learning which optimizes scheduling decisions
based on historical workload results |
 |
．Allow local cluster level optimization of most grid workload
．Enable broader resource access with automatic job translation which allows to effectively
run on more destination clusters |
| Grid Control with Automated Tasks, Policies and Reporting |
．Guarantee that the most important work is
prioritized first with flexible global grid
policies that respect local cluster policies and
support grid service level agreements
．Ensure availability of key resources and
specific times with advanced reservations
．Tune policies prior to roll out with cluster and
grid-level simulation
|
 |
．Use a global view of all grid operations for grid self-diagnostics, planning, reporting and
grid-wide and per cluster accounting across all resources, jobs and clusters
．Create Virtual Private Clusters so users only need to see the resources and jobs they have
access to without the complexity of seeing or wanting what is off limits |
| Controls for Cluster Sovereignty and Trusted Sharing |
．Guarantee that shared resources are allocated
fairly with global policies that fully respect local
cluster configuration and needs
．Establish trust between resource owners through
built in usage controls, accounting and
graphical reporting of usage across all shared
resources
|
 |
．Maintain cluster sovereignty with granular controls over where jobs can originate
and
where jobs can be processed
．Establish resource ownership and enforce priority access to resources when they want with
owner-based prioritization, preemption, and access guarantees |
| Increase User Collaboration and Productivity |
．Reduce end user training and job management
with easy-to-use graphical interfaces that
automate data migration and leverage unified
credential mapping
．Enable end users to easily submit and manage
their own jobs through a web browser to
minimize the costs of managing a growing base
of needy users
|
 |
．Collaborate more effectively with multi-cluster co-allocation of resources, allowing
compute, network, and data resources to be reserved for key projects
．Leverage saved job templates allowing users to quickly submit multiple jobs with
minimal
changes |
| Unify Management Across Diverse Independent Clusters |
．Unify management across existing internal,
external and partner clusters and their diverse
resource managers, databases, operating systems
and hardware
．Get out-of-the-box local area grid and wide area
grid support with scalability to more than 100
clusters and tens of thousands of diverse nodes
|
 |
．Manage secure access to resources with simple credential mapping or seamless integration
with Globus's advanced security toolset
．Leverage existing data migration technologies such as SCP, GASS or GridFTP |
*Operating system support for Linux (all), Unix (AIX, IRIX, HP -UX, FreeBSD, OSF/Tru-64,
Solaris,& other
UNIX platforms), Mac OSX & Windows.
**Resource Manager support for LSF, TORQUE, LoadLeveler, OpenPBS, PBSPro, BProc, SLURM, Warewulf & Native; Condor & SGE under development.
|
|
|
| ClusterResources |
 |
|
