- Non-virtualized, 'bare-metal' execution for optimal performance
- Low-latency QDR Infiniband and 10GbE fabric on all nodes for great application scalability
- Fast CPUs, GPUs, large memory, local scratch space, fast global storage space
- Over 60 HPC applications installed (FEA, CFD, Life Sciences, Weather forecasting)
- Flexible virtual server environment for job submission and hosted web applications
- Support from experienced HPC system administrators
- Pay-as-you-go only for core hours really used by your jobs - billing at 3-second intervals
You don't need to own a powerful cluster to run your HPC simulations at large scale.
With Penguin Computing on Demand (POD) you can use our HPC resources and pay-as-you-go. POD's compute environment was designed specifically for high-performance computing and features typical HPC components such as low-latency interconnects and GPUs. For optimum performance all jobs are managed by industry leading HPC schedulers that support the job submission semantics of the open source scheduler TORQUE. All jobs are executed directly on our HPC servers without a virtualization layer in the middle.
You access POD's HPC resources through a personal, virtual login node that is dedicated to you for the life-time of your account. All configuration settings are persistent while your account is active and there is no need to constantly allocate and ‘tear down’ compute instances. You submit your compute jobs from the login nodes through the command line or use the job submission and monitoring interface Scyld Insight. Once a month we bill you for the actual usage time you incurred and storage capacity you allocated. It's that easy.
Beyond offering an easy-to-use HPC compute environment POD provides centralized management capabilities through the POD portal. Multiple SSH keys, user accounts, login nodes and storage volumes can be managed from a central console enabling system administrators to grant and control access to POD resources.
All POD users also have access to PODTools, a freely available collection of applications that interface with POD directly from a user's personal machine for job submission, job queries and reporting without the need to log into a POD login node.