LSF Version 7.3 - Administering Platform LSF

About LSF on Platform EGO
186 Administering Platform LSF
About LSF on Platform EGO
LSF on Platform EGO allows EGO to serve as the central resource broker, enabling
enterprise applications to benefit from sharing of resources across the enterprise
grid.
◆ Scalability—EGO enhances LSF scalability. Currently, the LSF scheduler has to
deal with a large number of jobs. EGO provides management functionality for
multiple schedulers that co-exist in one EGO environment. In LSF Version 7,
although only a single instance of LSF is available on EGO, the foundation is
established for greater scalability in follow-on releases that will allow multiple
instances of LSF on EGO.
◆ Robustness—In previous releases, LSF functioned as both scheduler and
resource manager. EGO decouples these functions, making the entire system
more robust. EGO reduces or eliminates downtime for LSF users while
resources are added or removed.
◆ Reliability—In situations where service is degraded due to noncritical failures
such as sbatchd or RES, by default, LSF does not automatically restart the
daemons. The EGO Service Controller can monitor all LSF daemons and
automatically restart them if they fail. Similarly, the EGO Service Controller
can also monitor and restart other critical processes such as FlexLM and
lmgrd.
◆ Additional scheduling functionality—EGO provides the foundation for
EGO-enabled SLA, which provides LSF with additional and important
scheduling functionality.
◆ Centralized management and administration framework.
◆ Single reporting framework—across various application heads built around
EGO.
What is Platform EGO?
Platform Enterprise Grid Orchestrator (EGO) allows developers, administrators,
and users to treat a collection of distributed software and hardware resources on a
shared computing infrastructure (cluster) as parts of a single virtual computer.
EGO assesses the demands of competing business services (consumers) operating
within a cluster and dynamically allocates resources so as to best meet a company's
overriding business objectives. These objectives might include
◆ Reducing the time or the cost of providing key business services
◆ Maximizing the revenue generated by existing computing infrastructure
◆ Configuring, enforcing, and auditing service plans for multiple consumers
◆ Ensuring high availability and business continuity through disaster scenarios
◆ Simplifying IT management and reducing management costs
◆ Consolidating divergent and mixed computing resources into a single virtual
infrastructure that can be shared transparently between many business users
Platform EGO also provides a full suite of services to support and manage resource
orchestration. These include cluster management, configuration and auditing of
service-level plans, resource facilitation to provide fail-over if a master host goes
down, monitoring and data distribution.