It always surprises me when researchers say, “wait…Oracle has a Cloud?”
In fact, we do – a very robust and secure cloud with High-Performance Computing (HPC), fast networking, Autonomous Database, and more. Through Oracle for Research, researchers get both free access to Oracle Cloud and free technical collaboration to optimize the use of cloud to accelerate their research. Through Oracle for Research, I’ve had the great pleasure of working with Klara Jelinkova, CIO, and her wonderful team at Rice University.
Rice is leading the expansion of research in Houston, sparking innovation and growing national visibility and computational capacity. When Klara applied for a 2-year National Science Foundation (NSF) infrastructure grant to expand Rice’s HPC capacity for research, I was thrilled she invited Oracle for Research to collaborate with her. Under the grant, Rice would build out its on-premises data center, develop infrastructure to burst into a commercial cloud when needed, and share 20% of this goodness with the Open Science Group to advance research across the U.S. At Oracle, we were eager to advance her vision and the research mission at Rice, and supporting the NSF grant application to expand computational capacity was a natural fit. If awarded, we anticipated a close collaboration, with teams from Rice and Oracle working side by side.
This was 6 months B.C. (Before COVID). By the time Rice learned it had been awarded the grant, the world was in lockdown. Like everything else, our collaboration went virtual. We kicked off the project via Zoom in October, 2020, with the overarching objective of exploring the efficacy of cloud-bursting research workloads from the on-premises HPC cluster into Oracle Cloud. To explore this, the team had to enable Open Nebula to interface with Oracle Cloud Infrastructure (OCI).
Bursting from an on-prem environment into a commercial cloud for extra capacity sounds like it should be straightforward – just take the research workload that needs more capacity and move it between environments. In reality, this is a complex technical challenge. Research users, some of whom have little technical knowledge, need to be able to interface with the system to submit their workloads. Research workloads must be organized and prioritized within the on-prem environment. The on-prem environment must determine when and what type of extra capacity is needed, and be able to manage the workloads to optimize use of the cloud capacity. The research workloads that will move to the cloud must be transmitted and processed securely, efficiently and without disruption of the on-prem environment. Research results must be returned to the researcher the same way – securely, efficiently and without disruption of the on-prem environment. And, cloud capacity must automatically be spun up and spun down so research dollars are optimized.
In understanding, this, a project that sounded like it was going to be an easy, one-mile flat loop is suddenly seeming a lot more like climbing Mt. Everest. In both cases, though, the wisdom of Chinese philosopher Lao-Tzu holds true: a journey of a thousand miles begins with a single step.
At Oracle, our technical teams engage in complex collaborative projects with customers every day. They are experts in simplifying the process through sophisticated project plans, assigning project roles and tasks, organizing cadence calls, and writing and testing code, and communicating in technical shorthand via Slack. Oracle for Research brings these technical collaborations to research teams at no cost. Working with Erik Engquist, Director, Center for Research Computing, and his team at Rice, Oracle for Research and the OCI team plotted a map of milestones to move us forward, identifying a systems architecture and required capacity, integrating tools like Kubernetes and SlateCI for containerizing and orchestrating research workloads, and connecting and aligning the on-prem architecture with OCI. The work includes developing and delivering an auto-scalable IaaS GPU enabled Kubernetes solution and the seamless integration of SlateCI with the Oracle Kubernetes Engine. Together, this enables Rice to administer and scale multiple on-premises and OCI Kubernetes clusters from a single interface.
When Oracle for Research engages in research collaborations, we actively look for ways to develop community and help make connections between researchers. We also aim to build reusable images, tools, and documentation that can help researchers across disciplines accelerate their results. Sometimes, we find our way to interesting projects because researchers bring them to us.
That was the case here: we were especially excited about the opportunity to develop the OpenNebula API driver and contribute it to the open source community. OpenNebula is an open source platform to build and manage customized, enterprise cloud environments, and is used across the Rice University campus to make it easy for students, faculty and researchers to use Rice’s IT infrastructure. It is a critical component of our collaboration with Rice to help expand their research computing capacity.
“We are delighted that this project with Oracle has enabled Oracle and Rice to contribute this resource back to the OpenNebula community,” said Klara Jelinkova, CIO, Rice University. “The close collaboration between Oracle technical experts and my team has provided learning opportunities and enabled us to explore new and innovative technical solutions.”
The creation and donation of the OpenNebula API is an early and exciting mile marker in the current journey that Oracle and Rice are on together, with many more to come. We are grateful for the ongoing and committed collaboration of our colleagues at Rice.
For more information about Oracle for Research and how you can gain access to Oracle Cloud and technical collaborations to advance your work, please visit Oracle for Research.