Based on NVIDIA CUDA™ GPU architecture, the DALCO r16364gx3 rounds out the portfolio of DALCO GPU high performance computing solutions that give researchers, engineers and scientists the most powerful computing tools on the market to generate large-scale breakthroughs and bring innovations to life. Combined with a DALCO r16364gx3, an advanced 3U 16GPU cluster building block, and two 1U Intel™ or AMD™ host systems, the DALCO GPU ClusterX is the highest-density 5U 16GPU server solution on the market today. The system was designed to concentrate tremendous computing power in a simple, scalable format that brings high-availability with redundant components. It features an unprecedented level of ease of serviceability with front-loaded hot-pluggable GPU modules and high-manageability with built-in IPMI.

In terms of performance, the GPU ClusterX provides 16.48 teraflops of single precision performance in a single chassis and features ECC memory and 4x dual-socket Intel® Xeon® 5600 series CPU for the most demanding, mission-critical applications. When compared to other competing GPU solutions, the GPU ClusterX delivers 25% greater density, 31% lighter weight, and 11% fewer racks and switches. The GPU ClusterX expands on the innovative DALCO ClusterX Server Series to give customers maximum flexibility in how they build and scale data center infrastructures, whether their needs revolve around the highest possible density, fastest performance or best power efficiency.

Features:

• Supports 16 Tesla GPUs in 3U space
• Delivers 7,168 CUDA cores, 16.48 teraflops of single precision and 8.24 teraflops of double precision performance in a single chassis
• Powered by the massively parallel CUDA architecture, transforming a workstation to perform like a small cluster at 1/20th the power consumption and 1/10th the cost
• CUDA programming environment with broad support of programming languages and APIs, including C, C++, OpenCL, DirectCompute, or Fortran
• Offers protection of data in memory to enhance data integrity and reliability for applications. Register files, L1/L2 caches, shared memory, and DRAM all are ECC protected
• NVIDIA® Parallel DataCache™ Technology, accelerating algorithms such as physics solvers, ray-tracing, and sparse matrix multiplication where data addresses are not known beforehand
• NVIDIA® GigaThread™ Engine, maximizing throughput with faster context switching, concurrent kernel execution, and improved thread block scheduling Ideal for life sciences, geosciences, engineering & sciences, molecular biology, medical diagnostics, electronic design automation (EDA), government and defense, visualization, financial modeling, and oil & gas applications

[Contact us] for Informations or your individual quote