Open source private HPC cloud specialist vScaler today announced the integration of SLURM workload manager with GigaIO’s FabreX for elastic scaling of PCI devices and HPC disaggregation.
FabreX, which GigaIO describes as the “first in-memory network,” supports vScaler’s private cloud appliances for such workloads such as deep learning, biotechnology and big data analytics. vScaler’s disaggregated (composable) HPC solution is enabled by FabreX’s cloud infrastructure platform, a PCIe network that allows users to create rack-scale servers that can be reconfigured on the fly as workflows change, improving the utilization rate of expensive resources like GPUs and FPGAs, according to GigaIO.
The offering’s objective is to address problems related to static hardware configurations in HPC infrastructures. The companies said their “elastic cloud approach” enables users to “add a pool of movable resources into their overall architecture and then map those to the devices as and when they are required.”
“We’ve integrated GigaIO’s FabreX hardware into our private cloud product to allow our customer dynamically reconfigure hardware in line with evolving user demands and workloads,” said vScaler CTO David Power. “As people start to employ more AI tools and techniques, the underlying requirements for hardware to accelerate those tools also evolves. FabreX allows us provision resources to workloads so that we can run much more diverse workloads on top of a core hardware platform”.
The integration of SLURM, an open-source job scheduler for Linux and Unix-like kernels, means that vScaler Cloud users can request resources such as memory and compute cores for jobs. “Coupled with FabreX, users can now also specify PCI devices such as NVMe or GPUs to be attached to cluster nodes when running and executing specific workloads,” vScaler said in its announcement.
“FabreX is based on an open architecture offering industry standard Redfish APIs that make it easy to setup, and because we partner with top tier providers like vScaler for seamless integration, we’re delivering true cloud-class orchestration and composition capabilities,” said Alan Benjamin, CEO of GigaIO. “vScaler’s SLURM integration with FabreX enables end users to dynamically compose their own infrastructure by reaching inside the rack to create servers and resources in seconds, complete with leading security, access control and provisioning features, to match the needs of today’s workflows used in cloud deployments while optimizing TCO.”