Less Power is
the New Power
Fit more compute into your power and thermal envelope at the edge
Increase compute capacity and scalability in space and power constrained environments.
Accelerate your edge AI device—get more compute and extra room for GPU in your power budget.
Deliver predictably fast responses and freedom from interference on single threaded cores.
Move quickly and seamlessly from development to deployment of your embedded product with the ADLINK COM-HPC reference developer platform with reference carrier board—powered by Ampere.
“...perhaps the most powerful Arm-based machine that runs the consumer-oriented Microsoft operating system.”
Get the most powerful, energy efficient, and small form factor ADLINK COM module in the embedded market—scaling from 32 to 128 cores for intensive workloads and demanding hypervisors.
“Cloud-native app developers are getting a new tool to get the job done.”
Run your edge computing application on this Red Hat OpenShift certified cloud native platform optimized for Telco Edge, Open-RAN, and edge AI with optional NVIDIA GPUs and DPUs.
“... designed for an organization that wants a new Arm-based server with open technologies and non-proprietary building blocks.”
Get powerful, reliable compute in a compact, rugged design ideal for harsh environments— delivering edge compute with NVIDIA GPU for vehicles and other demanding applications.
“... seeing a 45W TDP 32 Arm core/ 128GB RAM machine is interesting, especially when it comes in a ruggedized platform.”
Watch our webinar to learn about the shift to cloud native solutions for edge computing and discover how you can achieve faster validation, higher performance, and increased efficiency.
Ampere Altra Family features 32 to 128 single threaded cores on every chip.
128 PCIe Gen4 lanes make room for GPUs, accelerators, storage, and other expansions.
Add Ampere Altra to your RPi and other Arm64 environments to scale up and scale out.
*The edge computing comparisons here are based on benchmark comparisons measured and published by Ampere Computing.
Details and device level footnotes are available here.