NetworkComputer is a high-performance enterprise grade job scheduler designed for distributed HPC environments. NetworkComputer provides a cost-effective highly adaptable solution capable of managing compute infrastructures from small dedicated server farms to complex distributed HPC environments.
NetworkComputer’s small memory footprint, sub-second dispatch latency and scalable architecture makes it an ideal solution capable of handling instantaneous workloads in excess of 2 million jobs and capable of scaling to a throughput exceeding billions of jobs per day.
A full-featured scheduler, NetworkComputer is equipped with a comprehensive set of policy management features including FairShare, Preemption and Reservations, allowing you to customize it to your organizational needs and maximize your compute infrastructure resource utilization and throughput.
For additional information about NetworkComputer click the tabs below.
Full-Featured High-Performance Scheduler
NetworkComputer is a full-featured yet versatile job scheduler architected for the high performance distributed HPC environments. A high capacity scalable architecture allows you to manage complex workload environment capable of scheduling billions of jobs per day.
- Sub-Second Latency for Best Performance & User Experience
- Intelligent Job Submission via Resource Profiling
- Scalable Small Footprint Architecture for Best Resource Utilization
- Ease of Management with Visual Workload Monitoring and Customizable Alerts & Notifications
- High Capacity Scalable Architecture Capable of Scheduling Billions of Jobs Per Day
- Manage Organizational Priorities with Comprehensive Policy Management Features
A full-featured, High-Performance job scheduler, NetworkComputer is designed to manage complex EDA workload environments. NetworkComputer’s scalable event driven architecture combined with its small memory footprint and sub-second dispatch latency accelerates design throughput and maximizes capacity utilization capable of running billions of jobs per day.
NetworkComputer’s “Job Profiling” provides users with greater visibility on the resources required by their jobs allowing them to keep track of CPU, memory and license usage over the lifetime of a job. This allows users to select queues with resource allocations that better fit their job profile for optimal resource utilization and throughput.
NetworkComputer is equipped with a comprehensive set of policy management features including: FairShare and Preemption. These capabilities provide management with essential controllability and observability tools to set and alter their policies in order to optimize their workload environment to their needs.
NetworkComputer enables the implementation of fair share policies for both hardware resources and software licenses to ensure that business-critical work is completed on time. The fair share rules are applied to the users competing for resources. If there is no competition, a single user is allowed to use all necessary resources required to complete all tasks in the minimum amount of time.
Job preemption and reservations can be used when one or more urgent jobs require licenses that are currently used by other less important jobs. Instead of waiting for the jobs to finish, NetworkComputer can suspend the running jobs and free the licenses. When the urgent jobs are done, the suspended jobs are resumed. There is no need to set aside licenses for urgent jobs, a strategy that leads to inefficient license utilization. Instead, with preemption, NetworkComputer can achieve 100% license utilization while servicing all classes of jobs.
NetworkComputer provides unique visualization and failure identification capabilities that provide “at a glance” views of jobs that have completed, failed, and/or waiting. This provides system administrators with greater productivity to easily identify any given job status while quickly analyzing the root cause of a failing job by bringing up relevant log files.
System administrators can easily configure NetworkComputer to send automatic alerts to the users or administrators when anomalous situations occur. This automatic notification reduces the effort required to keep the job scheduling and the whole system performing at a high level.