1U Dual CPU Intel Xeon, 2x 2.5", 12 DIMM, GPU Server (Up to 4 GPU) - 1029GQ-TRT-NEBS
Ultra High Density 1U Parallel Computing Server
Dual Intel 2nd Gen Intel® Xeon® Scalable Processors
Supports up to 4 NVIDIA GPU
2933†/2666 ECC DDR4 RAM installed; supports up to 3TB
Up to 4 storage drives; 2 are hotswappable
Includes 2000W redundant power supply
Intel Xeon Phi Coprocessor x100: Knight's Corner
Single programming model for all your code. The Intel Xeon Phi coprocessor gives developers a hardware design point optimized for extreme parallelism, without requiring them to re-architect or rewrite their code. No need to rethink the entire problem or learn a new programming model; simply recompile and optimize existing code using familiar tools, libraries, and runtimes.
Performance multiplier. By maintaining a single source code between Intel Xeon processors and Intel Xeon Phi coprocessors, developers optimize once for parallelism but maximize performance on both processor and coprocessor.
Execution flexibility. Designed from the ground up for high-performance computing. Unlike a GPU, a coprocessor can host an operating system, be fully IP addressable, and support standards such as Message Passing Interface (MPI). Additionally, it can operate in multiple execution modes.
"Symmetric" mode: Workload tasks are shared between the host processor and coprocessor
"Native" mode: Workload resides entirely on the coprocessor, and essentially acts as separate compute nodes.
"Offload" mode: Workload resides on the host processor, and parts of the workload are sent out to the coprocessor as needed.
Intel Xeon Phi x200 ARK Page
Intel Xeon Phi x100 ARK Page
See below for our full selection of Intel Xeon Phi processor and coprocessor systems.