The CAPE project develops computer architectures for efficient edge clouds.
Supporting AI-driven environments with networks of autonomous devices that maintain context and individual and shared state requires a powerful locally deployed edge cloud infrastructure. This allows devices to work together and in unison towards individual shared goals. To scale sustainably, this infrastructure must be delivered as a service and built on a configurable hardware foundation. Compute, memory, storage, and accelerators must dynamically self-configure to maximize utilization, minimize waste, and adapt capacity and heterogeneity to local needs, delivering cloud-class performance through distributed, fine-grained deployment.
The EU-funded CAPE (European Open Compute Architecture for Powerful Edge) project is addressing this gap by establishing edge micro data centers as a new composable building block of the edge-cloud continuum. CAPE combines open hardware, open source software, and open standards to enable flexible and efficient sovereign edge computing across Europe.
Edge hardware platform: configurable by design
CAPE reimagines edge servers as dynamically configurable pools of resources rather than fixed-function systems. The platform is designed as an autonomous, fine-grained edge micro data center that can operate independently while collaborating with neighboring nodes to form a distributed edge cloud fabric. You can flexibly combine compute, memory, storage, and accelerators to match workload requirements, significantly increasing utilization and reducing overprovisioning.
This project develops two complementary open hardware platforms based on COM-HPC technology: the Embedded High Performance Server (eHPS) and the Embedded Micro Data Center (EMDC). These platforms support heterogeneous computing nodes and target a wide range of edge scenarios, from industrial on-premises deployments to communications environments.
Composable hardware is enabled by combining PCIe switching with Compute Express Link (CXL) to provide cache-coherent connectivity between processors, accelerators, and pooled memory. This allows each autonomous edge unit to adapt to fluctuating AI workloads while maximizing hardware utilization as resources are dynamically pooled and reallocated at runtime.
Energy efficiency is a core design goal. CAPE minimizes over-provisioning and unnecessary energy consumption by maximizing hardware utilization and reducing stranded resources across distributed deployments. Passive cooling, waste heat recovery, and building-level integration further reduce the environmental footprint of embedded edge infrastructure.
Software stack: Orchestrate the edge-cloud continuum
To take full advantage of the configurability of the hardware, CAPE is developing an open source software stack that hides the complexity of the infrastructure while maintaining control and transparency. This stack combines Kubernetes-based multi-cluster orchestration with AI-assisted automation. Orchestration maintains local autonomy and allows each edge unit to operate independently while participating in the coordinated coordination of the entire distributed infrastructure.
The core innovation is Infrastructure from Code (IfC). It derives infrastructure definitions directly from application code, reducing development effort compared to traditional infrastructure-as-code approaches. Applications can be seamlessly deployed across CAPE Edge servers and external cloud environments, allowing for flexible scaling and workload migration.
Unified management is provided by openMPMC, an open multiplatform management controller that provides web-based operations and Redfish-compatible APIs for configuration, monitoring, and dynamic resource configuration.
Standardization and openness
CAPE is committed to open standards and open source development. The project builds on technologies such as COM-HPC and CXL and actively contributes to European and international initiatives, avoiding its own lock-in and fostering a sustainable ecosystem for edge computing. This openness directly supports Europe’s goals for digital sovereignty across the edge-cloud continuum through the Open Federated Edge-Cloud Infrastructure Alliance.
Use case: Real-world environmental impact
CAPE validates its architecture through three representative use cases. In smart grids, edge servers enable real-time monitoring and anomaly detection at substations. For deep learning at the edge, heterogeneous accelerators and CXL-based memory pooling support efficient AI inference under tight power constraints. For satellite payload data processing, CAPE enables near real-time analysis through dynamic resource configuration across the edge and cloud.
outlook
CAPE will demonstrate how composability, open standards, and intelligent orchestration can transform edge computing into a flexible, energy-efficient foundation for future digital services. This project lays the foundation for a scalable sovereign edge infrastructure in Europe by tightly integrating hardware and software innovations.
Disclaimer
The CAPE project is funded by European countries.
The Union’s Horizon Europe research and innovation program was co-funded by the Swiss National Science Foundation (SNSF, number 200429) and Armas Swiss Science and Technology under grant agreement number 101189899.
Please note: This is a commercial profile
This article will also be published in the quarterly magazine issue 25.
Source link
