Creating a Dynamic HPC Infrastructure with Platform Computing

The Business Problems 2 Rigidity Inefficiency Incapacity Workload requirements are dynamic and only becoming

Platform’s Roadmap to HPC in the Cloud Efficiency, Flexibility, Service Levels Public Cloud Private

Typical Solutions Static infrastructure = trade-offs Utilization Static HPC Capacity Large Job Starvation Application

Introducing Platform Adaptive Cluster Utilization With Platform Adaptive Cluster, you can avoid this trade-off

Platform Adaptive Cluster Transform a static cluster into a fully dynamic cloud environment Increase

Create a Dynamic HPC Environment Dynamic. Consolidation Memory Provisioning of OS RHEL 5. 5

Use Case: Large EDA Customer Problem • Service level requirements for high priority workload

Any Questions? Chris Porter cporter@platform. com 9 Copyright © 2011 Platform Computing Corporation. All

Slides: 9

Download presentation

The Business Problems 2 Rigidity Inefficiency Incapacity Workload requirements are dynamic and only becoming more so Unbalanced utilization of distributed resources (hot and cold spots) Peak demand occasionally exceeds local AND total supply of resources Copyright © 2011 Platform Computing Corporation. All Rights Reserved.

Platform’s Roadmap to HPC in the Cloud Efficiency, Flexibility, Service Levels Public Cloud Private Cloud Grid 4. Make infrastructure dynamic within a shared private cloud Cluster 3. Make infrastructure dynamic 2. Employ effective resource management and sharing 1. Harness the power of commodity clusters 3 5. Burst to external providers when needed Coming soon! Platform LSF + Adaptive Cluster We Are Here Time Copyright © 2011 Platform Computing Corporation. All Rights Reserved.

Typical Solutions Static infrastructure = trade-offs Utilization Static HPC Capacity Large Job Starvation Application Stack Silos Queue Sprawl Cluster Sprawl Service Level Copyright © 2011 Platform Computing Corporation. All Rights Reserved.

Introducing Platform Adaptive Cluster Utilization With Platform Adaptive Cluster, you can avoid this trade-off Delivers both better economic value and service levels! Service Levels / Qo. S 5 Copyright © 2011 Platform Computing Corporation. All Rights Reserved.

Platform Adaptive Cluster Transform a static cluster into a fully dynamic cloud environment Increase user service level Redeploy servers quickly & efficiently Automate administration 6 • Reduce complexity of user environment • Control resource allocation policies at the group level • Benefit from a mature, stable HPC cloud product • Reduce cost & save power • Increase resource sharing & utilization • Achieve self-service for users • Allows the environment to flex as requirements change over time Copyright © 2011 Platform Computing Corporation. All Rights Reserved.

Create a Dynamic HPC Environment Dynamic. Consolidation Memory Provisioning of OS RHEL 5. 5 RHEL 4. 8 Big Mem Job Platform LSF + Adaptive Cluster Workload out of balance with OS provisioning Large memory jobs starved by running jobs RHEL 5. 5 4. 8 7 RHEL 4. 8 5. 5 RHEL 5. 5 Copyright © 2011 Platform Computing Corporation. All Rights Reserved.

Use Case: Large EDA Customer Problem • Service level requirements for high priority workload conflicts with the need to keep utilization high • Resources are reserved for critical workload • When there is no priority workload, utilization is low Alternatives • Use reserve resources, force priority workload to wait • Preempt low priority workload, kill or wait for priority Solution: Platform LSF + Platform Adaptive Cluster • Use reserved infrastructure for low priority workload • When priority workload arrives migrate low priority to other resources • Utilization is high, no workload is starved or lost 8 Copyright © 2011 Platform Computing Corporation. All Rights Reserved.