NextIO, a provider of next-gen I/O consolidation offerings, has introduced the vCORE Express 2090 GPU computing offering, which is based on the new NVIDIA Tesla M2090 GPUs.
The new NextIO vCORE Express 2090 enables customers to run GPUs outside of their servers, providing dense server configurations and lifecycle management.
The new offering is a high-performance computing server that is designed for demanding HPC applications. Servers powered by the vCORE Express 2090 deliver up to 10x higher application performance than the new multi-core CPUs, the company said.
The vCORE Express 2090 is an ideal offering for real-world applications such as seismic processing, biochemistry simulations, weather and climate modeling, signal processing, computational finance, CAE, CFD, and data analysis, said the company.
The vCORE Express 2090 capabilities include: 2048 computing cores for accelerated scientific computations (512 cores per GPU); 2.6 teraFLOPS for twice the precision performance (665 gigaFLOPS per GPU); ECC memory for uncompromised accuracy and reliability; and an L1/L2 cache for improved bandwidth and data sharing.
In addition, the offering is designed for parallel computing applications using NVIDIA Tesla 20-Series GPUs in cluster and data centre deployments.
NextIO sales and marketing senior vice-president Mike Heumann said the company’s goal is to deliver the highest performance I/O offerings to their partners and customers.
"Adding the vCORE Express 2090 GPU system to our product family provides enterprise and academic customers with the industry’s highest performing GPU offering," Heumann said.