Seneca Launches HPC Compute Nodes with New NVIDIA Tesla K20 GPU Accelerators

Share Article

Packing one petaflops in 10 racks with better power efficiency

Seneca®, a leading manufacturer and distributor of innovative technology solutions, announced today the availability of the Nexlink® HPC compute nodes powered by the new NVIDIA® Tesla® K20 family of GPU accelerators. Nexlink HPC systems are available in a series of rack configurations from 1U, 2U and blade rack elements designed to meet the needs of a range of clients, from small departmental research clusters to the highest density national labs.

Announced today, the NVIDIA Tesla K20 family of GPU accelerators delivers three times the performance per watt of previous-generation GPUs, accelerating a broad range of scientific, engineering, and commercial high performance computing (HPC) and data center applications.

The Tesla K20X GPU accelerator, the new flagship of NVIDIA’s Tesla product line, provides the industry’s highest single and double precision peak floating point performance—3.95 teraflops and 1.31 teraflops, respectively. The Tesla K20 GPU accelerator provides 3.52 teraflops of single precision and 1.17 teraflops of double precision peak performance. Based on the revolutionary NVIDIA Kepler™ compute architecture, the NVIDIA Tesla K20 family of GPU accelerators enables Nexlink HPC solutions to increase GPU utilization dramatically with greater power efficiency.

They include innovative technologies like Dynamic Parallelism and Hyper-Q to boost performance and deliver record application speeds for seismic processing, biochemistry simulations, weather and climate modeling, image, video and signal processing, computational finance, computational physics, CAE, CFD, and data analytics.

The innovative Kepler compute architecture design includes:

SMX (streaming multiprocessor) design that delivers up to three times the performance per watt compared to the SM in the earlier Fermi architecture. It also delivers one petaflop of computing in just 10 server racks.

Dynamic Parallelism enables GPU threads to spawn new threads automatically. By adapting to the data without going back to the CPU, it greatly simplifies parallel programming. Plus it enables GPU acceleration of a broader set of popular algorithms, like adaptive mesh refinement (AMR), fast multipole method (FMM), and multigrid methods.

Hyper-Q feature that enables multiple CPU cores to utilize the CUDA cores on a single Kepler GPU simultaneously. This dramatically increases GPU utilization, slashes CPU idle times, and enhances programmability, especially for cluster applications that use MPI.

Stop by Seneca’s booth #2334 at SC12 in Salt Lake City November 12-15th 2012. For more information about pricing and availability contact Seneca at (800) 227-3432 or visit

About Seneca:
For over 30 years, Seneca has met the needs of customers in diverse industries with original and partnered technology solutions. As a leading U.S.-based Custom Computer Manufacturer, Seneca serves the evolving needs of our Business, Gov/Ed, Digital Signage, Digital Security and Surveillance, Digital Health and Digital Broadcast customers nationwide. By establishing the highest standards for our products, we forge long-term, mutually beneficial partnerships that deliver results. Seneca is best known for our rigorously tested Nexlink brand of custom-built desktops, servers, notebooks and storage devices. Our success is built on service and consistent performance at every stage of the extended technology lifecycle. For more information visit

Seneca. Creating & Partnering Throughout the Technology Lifecycle.

# # #

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Sarah Colson
Follow us on
Visit website