Engineers at every company are trying to innovate faster while holding down costs. Modeling and engineering simulations are the backbone of these efforts. Engineers may wish to run ANSYS Fluent simulations at scale, or many different permutations simultaneously, that may require more computing resources than are readily available. Hybrid HPC computing combines public and on-premise compute resources to offer organizations a flexible, cost-effective approach to meet these requirements.
To fully augment on-premise HPC systems requires cloud platforms that can scale-out as bare-metal environments. While moving all compute workloads from on-premise to cloud and back may be problematic for some, dividing workflows into on-premise “core” compute while using cloud resources for “project” workloads can be helpful. Leveraging a HPC cloud can provide organizations with a variety of benefits, ranging from economic ones to access to advanced technologies or temporarily increased capacity. A high-performance computing cloud as part of your infrastructure might also make sense even if the economics don’t.
The Value of a Hybrid Approach
HPC clouds provide access to newer technologies than a traditional three-year hardware refresh cycle would. Additionally, the right cloud provider can grant access to HPC expertise that will augment your existing IT team. To make the most out of hybrid computing, balancing the size of your on-premise systems to manage “core” workloads while using an on-demand HPC cloud to run schedule-sensitive “project” workloads can be the most economically optimized approach.
The Importance of Virtual Workstations
A key technology for hybrid cloud or for using any HPC cluster remotely is virtual workstations. Virtual workstations solve two problems. One is data locality. Simulations can produce large amounts of raw output data that is traditionally downloaded to a local workstation for post-processing. The time spent waiting for the download is unproductive and disrupts a smooth workflow. By using a virtual workstation, post-processing can be performed where the raw output resides, making it possible to download only the desired results.
The second way virtual workstations enable hybrid HPC is by providing a familiar environment for users. ANSYS has long integrated many tools in their ANSYS Workbench simulation platform to improve engineers’ agility and effectiveness, including seamless submission of simulations to a HPC cluster. However, making the connection between local workstations and a remote cluster can often prove difficult, if not impossible, due to security or efficiency concerns.
The ability to run Workbench on the same high speed network as the one that handles data and compute preserves the advantages of this comprehensive solution.
Note that all virtual workstations are not created equal. Running Workbench might require a simple 2-D desktop, while CFD-Post will require a higher powered 3-D workstation. Be sure to run tests to determine the tools you need before committing to a platform.
Comparing the cost of on-premise and HPC cloud computing involves a number of factors, but perhaps the most significant is utilization. Utilization is the percent of time a compute resource (core, processor, server) is in use by the application. Cost-effective HPC clouds only charge for the core-seconds that an application uses, along with the cost to store user data. On-premise costs include the capital expense of the cluster, network and storage divided by the utilization plus operating costs (IT staff, power, cooling, facility). Designing an on-premise cluster to meet peak demands without incurring application wait times for the users can mean that utilization will fall — often to the point where an HPC cloud will be more cost-effective. Balancing core compute requirements and flexible project-based compute requirements leads to an optimal solution.
Watch this on-demand webinar to learn how CRB, a leader in engineering, construction, architecture and consulting, is using ANSYS on Penguin Computing's public cloud, Penguin on Demand, to solve pharma and life science design challenges.
Penguin Computing’s Hybrid Cloud for ANSYS
Penguin Computing, a leading U.S. developer of open Linux-based HPC, enterprise data centers and cloud solutions has supported ANSYS users on its public cloud, Penguin Computing on Demand (POD), since its inception in 2010. Backed by true HPC bare-metal clusters with InfiniBand high-speed parallel file systems and the latest CPU models, POD also has an easy-to-use cloud-based portal. All users can access a free desktop capable of running Workbench with GPU-backed workstations for demanding post-processing, all though a standard web browser.
With very competitive pricing, world class support, unmetered free bandwidth and a full range of ANSYS software available, engineers overwhelmingly pick POD for their simulation needs.