Intel Xeon 6766E Processor (Sierra Forest, 144 E-cores)

The Intel Xeon 6766E processor, part of the Xeon 6 family built on the Sierra Forest microarchitecture, delivers 144 efficiency-cores designed to maximize compute density within a constrained power envelope. Fabricated on Intel's 3nm-class Intel 3 process node, this processor targets cloud service providers and enterprise data centers that require high thread counts per socket and predictable per-core performance for scale-out workloads. It supports a single-socket configuration and is engineered to provide substantially more cores per rack unit than prior-generation Xeon Scalable platforms.

The Xeon 6766E integrates support for DDR5 memory across eight memory channels, enabling high aggregate memory bandwidth to feed its large core count. The processor supports Intel's advanced RAS (Reliability, Availability, and Serviceability) features, hardware-enforced security capabilities including Intel Total Memory Encryption-Multi Key (TME-MK), and is compatible with platforms designed around the LGA 4710 socket. It carries a 330W TDP, reflecting its dense core configuration while remaining within data center power planning norms for high-core-count efficiency processors.

Targeted at horizontally scaled cloud infrastructure, containerized microservices, and telco network function workloads, the Xeon 6766E is purpose-built for environments where workload density and operational efficiency per watt matter more than single-threaded peak performance. Its E-core architecture omits Hyper-Threading in favor of native physical cores, delivering a high core count that benefits embarrassingly parallel and lightly-threaded-per-instance workloads common in modern cloud-native deployments.

Ideal for

• High-density cloud compute nodes running containerized microservices and Kubernetes orchestration at scale
• Telco and network function virtualization (NFV) deployments requiring many parallel lightweight processing threads per socket
• Web serving and front-end application tiers where maximizing concurrent session handling per rack unit is critical
• Big data analytics pipelines and distributed processing frameworks such as Apache Spark benefiting from high core parallelism
• Software-defined networking and packet processing workloads that scale horizontally across many cores
• HPC pre/post-processing and parameter sweep jobs where high thread counts reduce time-to-result across large job queues

Technical specifications

Available from Omnixon Global. Submit an RFQ and our team will confirm configuration and availability for your order.