Dedicated Hardware Solutions

Dedicated Hardware Solutions

Cirrascale Cloud Services was born out of a hardware company that was dedicated to engineering solutions that were the talk of the HPC industry. Early on, we leveraged our engineering resources and award-winning intellectual property to help our customers find the right deep learning training infrastructure for larger, more accurate neural networks. We worked closely with our customers to create and tailor custom rackmount solutions that met their specific needs.

Today, that same hardware infrastructure is the backbone of the Cirrascale Cloud Services platform and has been growing and improving from the years of valuable feedback received from our customers. In May 2017, BOXX Technologies, the leading innovator of high-performance computer workstations, rendering systems, and servers, announced the acquisition of Cirrascale Corporation. The acquisition enabled BOXX to add Cirrascale's deep learning hardware to its line of multi-GPU solutions and solidified BOXX as the leader in multi-GPU computer technology. Cirrascale Cloud Services now continues on to expand its growing multi-GPU cloud business and provide GPU-as-a-Service and other professional services offerings.

The Most Robust Deep Learning Hardware

Over the past several years, we have worked with a variety of application developers as well as data scientists in both industry and academia. Together, our goal has been to continue making groundbreaking improvements in developing some of the most advanced hardware capable of increasing the overall speed and flexibility of deep learning training and inference by using multi-GPU compute solutions.

By utilizing the BOXX Technologies SR3615 PCIe 96-lane switch riser, the newly designed BOXX GX8 supports up to eight NVIDIA Tesla V100 32GB GPU cards and provides room for additional InfiniBand® or NVMe storage devices while enabling increased bandwidth and lower latencies between PCIe Gen3 devices than are possible in traditional systems. By enabling up to 8 discrete GPU accelerators to communicate directly with each other on the PCI bus, free of the need for host CPU intervention, they can create a "micro-cluster", sharing a single memory address space.

Learn More About
Dedicated Hardware

Interested in learning more about dedicated deep learning workstations and servers? Visit BOXX Technologies for more information.