Dedicated Computing, a leader in embedded computing solutions for mission-critical applications, today announced the launch of two new Intel® Xeon® 6-based commercial rack servers: the Excalibur™ E6710 (1U) and Excalibur™ E6740 (4U). These platforms mark the expansion of Dedicated Computing’s commercial server series, delivering enterprise and HPC-class performance for customers accelerating AI, virtualization, GPU consolidation and high-bandwidth I/O workloads.
Complementing Dedicated Computing’s long-lifecycle embedded portfolio, these systems are purpose-built for customers who require standard server chassis options, very high computing performance, and flexible GPU or I/O configurations aligned with enterprise CPU roadmaps.
As medical imaging, life sciences, training & simulation, and advanced analytics environments rapidly increase compute density, customers are seeking standard rack server platforms capable of delivering extreme performance in condensed, rack-optimized systems. The Excalibur series addresses this demand with dual-socket Intel Xeon 6 performance, PCIe Gen5 expansion and accelerator-ready architectures.
“The Excalibur E6710 and E6740 are built for customers pushing AI and HPC performance in medical, life sciences, and advanced compute environments,” said David Galus, Director of Product Marketing at Dedicated Computing. “These platforms deliver the density, accelerator flexibility and next-generation Intel performance required for modern AI infrastructure.”
Excalibur E6740: 4U AI Infrastructure Platform with Up to 8 GPUs
The Excalibur E6740 is a dual-socket 4U rackmount server designed for enterprise AI infrastructure and high-performance computing. Powered by 6th Gen Intel® Xeon® 6 processors, the platform supports up to eight double-wide 600W GPUs. With 32 DDR5 DIMM slots supporting up to 4TB of memory, eight front hot-swap NVMe bays, and 3+1 redundant 3200W Titanium power supplies, the E6740 is engineered for sustained AI training, GPU consolidation, model inference clusters, and data-intensive HPC workloads.
Configurable NVIDIA NVLink bridge support enables high-bandwidth GPU scaling for demanding AI and parallel compute environments. The system is targeted at customers building AI servers requiring the highest levels of performance and maximum accelerator density.
Excalibur E6710: Dense 1U Dual-Socket Compute for Medical and Life Sciences
The Excalibur E6710 delivers dual-socket Intel Xeon 6 performance in a compact 1U form factor. Designed for medical and life sciences use cases, it supports 32 DDR5 DIMMs (up to 4TB) and 12 front 2.5-inch hot-swap bays supporting NVMe, SATA, or SAS storage. Its front-parallel CPU and memory architecture enables efficient airflow and excellent thermal performance, optimizing the platform for dense rack deployment. With up to four PCIe Gen5 expansion slots, including dual x16 links, the E6710 supports accelerator cards, high-speed networking, and storage controllers within a condensed footprint.
The platform is ideal for genomics processing, imaging back-end compute, FPGA-accelerated workflows, virtualization nodes, and AI-enabled analytics in space-constrained environments.
Commercial-Class Expansion of the DC Portfolio
The Excalibur series complements Dedicated Computing’s long-lifecycle embedded platforms by introducing high-performance commercial server options aligned with enterprise CPU roadmaps. These systems are intended for customers prioritizing extreme compute density, GPU scalability and flexible I/O in standard rack chassis.
Both platforms support Windows Server and Red Hat Enterprise Linux environments, with planned NVIDIA and Red Hat ecosystem certifications.
The Excalibur Series E6710 and E6740 are available now.
© 2026 Dedicated Computing LLC – All Rights Reserved. All product and company names are trademarks ™ or registered trademarks ® of their respective holders. Intel, the Intel logo, Xeon, and vPro are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others.