Welcome, Please Sign In

Get in touch with your rep, view past orders, save configurations and more. Don't have an account? Create one in seconds below.

login

AI/HPC Edge Reference Architectures

Accelerate workloads despite form factor and field limitations

Edge AI/HPC Reference Architectures

The plethora of data available today – and the ever-increasing rate of new data being created – and improvements in DevOps has led to the creation of many powerful new algorithms to extract more value from data.

But implementing HPC and AI on the edge requires a strong understanding of both computing and form factor. That’s why Silicon Mechanics created a series of reference architectures for specific types of edge deployment and workloads.

Each one is the result of hours of engineering, testing, and optimization for power, latency, and related concerns of resource-hungry applications as well as space, size, ruggedization, and other issues facing in-field deployments. That’s because we want to save clients time and focus on customizing the design for your specific workload and organizational needs –not redoing the basic elements with each new engagement.

Each reflects all our past work designing edge devices to meet the unique demands AI and HPC places on hardware. And they are a great starting place for your customized edge deployment.

Learn more about our edge HPC and AI reference architectures:

Argos Ruggedized Edge Appliance

Devices that support AI on the edge (i.e., such as in-vehicle or MIL-SPEC systems) are complex to design and usually only support a specific, brand-name edge cluster. The Argos Ruggedized Edge Appliance includes commodity hardware nodes designed to resist environmental conditions and powerful enough to support accelerated computing, without the costs of brand-name devices.

View
Prometheus Edge Appliance

Some organizations need cloud-like services (i.e., storage, compute, containers, GPU) in edge environments. But brand-name solutions like the AWS Snowball are expensive and have vendor-specific configurations. The customizable Prometheus Edge Appliance (codenamed "fireball") combines ruggedized, enterprise-grade components and a private cloud software stack suitable, all designed for use in the field, for the same performance but lower cost than Snowball.

View
Titania CDI Edge Cluster

Running a cluster outside of a datacenter can limit the workloads you can support. You have limited footprint, environmental concerns (vibration, heat, dust, moisture), power envelope limitations and other issues. The Titania CDI Edge Cluster gives you the power to reconfigure resources on the edge and still get bare metal performance. At the same time, it addresses form factor and deployment limitations you find at the edge.

View
Umbriel HCI Edge Cluster

A secure enclave/cloud environment is complex to design. Putting that capability at the edge makes it even more challenging and costly. The Umbriel Edge HCI Cluster solves the problem with a dense, scalable central cloud with a smaller, localized cloud-in-a-box at the edge, all using commodity components to keep costs down.

View

Expert Included

Our engineers are not only experts in traditional HPC and AI technologies, we also routinely build complex rack-scale solutions with today's newest innovations so that we can design and build the best solution for your unique needs.

Talk to an engineer and see how we can help solve your computing challenges today.