WorkloadXelerator is a high-performance workload specific scheduler designed for today’s distributed high-performance-computing (HPC) environment. System on chip (SoC) design workloads and throughput requirements are increasing particularly in simulation and characterization as demand increases for higher job counts and faster turn-around-time.

WorkloadXelerator is architected to off-load the base scheduler, enabling greater throughput, better license and resource utilization, and more flexible scheduler usage models.

WorkloadXelerator works with NetworkComputer™, IBM® Platform™ LSF® and other third party schedulers.  The upper level scheduler(s) can be designated for specific workloads, teams or individual users.

For additional information about WorkloadXelerator click the tabs below.

 

 

For Designers

  • Exclusive access to “private” scheduler, allows “self-directed” flows
  • High Throughput for design verification and IPQA runs

 

For Infrastructure

  • Flexible Usage Models (On-Demand, Workload, Project)
  • 6-10X Higher Throughput without adding compute capacity

For Management

  • Improved License Management
  • Fairshare and Preemption
  • Global Policy and License Management

Increased Throughput

Job Profiling

WorkloadXelerator incorporates the same event-driven core architecture as NetworkComputer, featuring millisecond dispatch latency for job scheduling, yielding the most benefits for short jobs. It serves as a “personal schedulers” for designers, allowing them to queue short jobs sequentially on their own.  This reduces the scheduling burden on the lower level scheduler by presenting a batch of short jobs as one longer job while maintaining full visibility into each individual job.

WorkloadXelerator schedulers also handle all the user query, job submission and reporting functions. The offload from the base scheduler is substantial since 80% of typical scheduler loads come from non-dispatch functions.

For highly parallel, short regression jobs with consistent license and computing resource needs, WorkloadXelerator deployments realize a 6-10X increase in throughput with no added compute capacity.

While WorkloadXelerator is not restricted to particular flows, those typically seeing the largest benefit include design verification regressions, IPQA and library characterization.

 

Greater Performance and Visibility

WorkloadXelerator offloads 3 high throughput jobs from NetWorkComputer Base Scheduler

WorkloadXelerator is also equipped with a comprehensive set of policy management features including Fairshare and Preemption. Cross-organizational resources can be assigned and managed as users prefer; by workflow, project or on-demand.

The full-featured GUI enables full visibility into each job (Running, Queued, Completed, and Failed) and compute farm statistics for Queued Jobs, Running Jobs, and allocated resources. Additional interfaces are web browser, command line and a TCL API.

Under a typical high-load situation of two regressions and a smoke test, WorkloadXelerator manages the peak loads, and even when combining all 3 workloads the effective load on the base scheduler stays stable at a level that is less than any of the individual workloads.

 

 

Datasheets

WorkloadXelerator Datasheet

White Papers