- Ease of Use - compute resources are available immediately
- Access to the latest HPC technology
- Outstanding Technical and Applications Support included
- Scalable computing resources
- Many commercial and open-source applications available
- multiple PODs available, each suited to different workload styles and industries
- Only pay for what you need and use, just core hours and storage - that's it!
Penguin's On-Demand HPC Cloud Service (POD). You don't need to own a powerful cluster in order to perform HPC tasks. POD users have a persistent and secure compute environment that executes jobs directly on the compute nodes' physical, not virtualized, cores for true HPC. Our POD technical team provides expert-level guidance and support to help you achieve high performance results.
Through your personal dedicated login node, POD provides a persistent Linux environment that includes a complete HPC software stack and is directly connected to a tightly-coupled, non-virtualized compute cluster for job execution. POD's compute cluster was designed specifically for high-performance computing and features typical HPC components such as low-latency interconnects, GPUs and high-performance parallel file systems. POD is cost-effective, convenient and secure, plus you only pay for the actual computing resources that you use.
Our lab at UCSF performs genetic research, studying how innate differences in everyone's DNA affect our susceptibility to skin and breast cancer. I had previously used another cloud service provider to perform the intensive statistical analysis required in our research, but it was a chore to manage twenty servers at a time. It has been a pleasure working with Penguin's POD. It's trivial to scale up as many cores as I need at a significant cost savings... I can run more experiments in less time at less cost with less hassle. My contact in Penguin Professional Services has been more like a collaborator than a vendor, and his observations about my system use identified a major inefficiency in my code. Fixing this resulted in a four-fold decrease in processing time. Analyses that would have taken months using my previous methods can now be done in a day or two.