ThunderX Based ARKA Platform
ARKA™ Platform for Specific Workloads in HPC, Cloud and Big Data Applications ARM-based ARKA servers  
News & Events > Press Releases   
64-bit ARMv8-A server processor  
E4 Computer Engineering Announces ThunderX™ Based ARKA™ Platform for Specific Workloads in HPC, Cloud and Big Data Applications

SAN JOSE, Calif., March 17, 2015 — E4 Computer Engineering a leading Italian manufacturer of hardware solutions for high performance computing and data center today announced a new series of ThunderX ARMv8-A enabled ARKA dense computing sever product line. The new Cavium ThunderX™- based systems leverage the design of the proven and successful ARKA modular servers, which are well positioned to handle the requirements of specific workloads in the Cloud, Big Data and HPC applications.

This new series of ThunderX based ARKA servers ensure optimal power efficiency and represents the best-in-class solution.  While previous ARM-based technology products were limited to 32 bit and were used only by early adopter and developer, the ARKA Series is the first production ready 64-bit ARM server platform that integrates ThunderX along with Graphics Processing Unit (GPU) and InfiniBand (IB). The ARKA series is equipped with full support of the extensive software ecosystem including Linux OS, libraries, compilers and applications. It is also equipped with the following development tools: NVIDIA CUDA 6.5 (compilers, libraries, SDK) MPI libraries, GNU compilers.

The ThunderX product family is Cavium’s 64-bit ARMv8-A server processor for next generation Data Center and cloud applications. With high performance custom cores, single and dual socket configurations, high memory bandwidth, large memory capacity, integrated hardware accelerators, integrated feature rich high bandwidth network and storage IO, fully virtualized core and IO, and scalable high bandwidth, low latency Ethernet fabric ThunderX enables best in class performance per dollar and performance per watt. The ThunderX processor family is fully compliant with ARMv8 architecture specifications as well as ARM’s SBSA and SBBR standards and is widely supported by industry leading OS, Hypervisor and SW tool and application vendors.

“We are very happy with the collaborative work done together with Cavium to develop the ARKA platform and are confident that the route to low-power efficiency for HPC is getting paved,” said Piero Altoè BDM E4 Computer Engineering.

“ThunderX powered ARKA Series server provides a versatile and flexible platform for the hyper-scale data center,” said Rishi Chugh, Director, Data Center Processor Group at Cavium. “E4 Computer Engineering extensive experience in HPC solution will enable highly competitive ThunderX based ARKA servers that are optimized for the Cloud, Web 2.0, HPC, and High Density Storage.”

“We are excited to see E4 Computer Engineering launching their first workload optimized ARM based server platform. This will drive momentum and further accelerate the deployment of ARM servers in the HPC and Cloud Infrastructure,” said Filippo Mantovani, Senior Researcher, Barcelona Supercomputing Center.

The ARKA systems will be showcased at GTC, the GPU Technology Conference held in Silicon Valley from March 17th – 20th, at E4 Computer Engineering booth 430 and Cavium booth 511.

About E4 Computer Engineering
E4 Computer Engineering spa specializes in the production and architectural integration of highly customized HPC & Data Center solutions for scientific/research and Enterprise environments. From designing HPC clusters to implementing turn-key data centers, E4 guarantees quality and expertise by consistently investing in R&D to meet the ever changing requirements of customers.

Media Contact
Ludovica Delpiano
Marketing Department
E4 Computer Engineering Spa
Direct line: +39 0522 991816
Cell: +39 346 2161214




ThunderX™ is a registered trademark of Cavium, Inc. All other trademarks are the property of their respective owners.

64-bit ARMv8-A Cpu

All contents are Copyright © 2000 - 2017 Cavium. All rights reserved.     Privacy Policy   |   Copyright Policy   |   Site Map