IDT Treats Network and Data Center Operators with a Vibrant Mobile Edge Computing Platform

By CIOReview | Wednesday, October 21, 2015
467
742
165

SAN JOSE, CA: With the increasing plethora of unstructured data, the complexities related to it are also increasing. But until now, as network operators have found out a nimble way to add analytic functionalities on the colossal of data storage, Integrated Device Technology (IDT) has rolled out a low-latency computing platform to speed big data analytics and deep learning at the edge of networks.

The mobile edge diverse platform heads straight way from IDT’s Open High-Performance Analytics and Computing (HPAC) lab, and is brought in to action to fuel up computing services in access networks and video distribution. However, the platform utilizes modules under beta tested conditions and draws its innovations in partnership with NVIDIA, Prodrive Technologies and Concurrent Technologies..

IDT provides the network and data center operators in association with system integrators, with an agile methodology to compute at the edge by linking accelerators based on a disparate architecture with the company’s low-latency RapidIO products.

The newly versed platform is built on 1U form factor server and features heterogeneous computing with NVIDIA Tegra K1 processors and x86 CPUs. Whereas, the core part of the product is connected with RapidIO fabric technology with about 100ns switching latency and 20 Gbps ports, while the futuristic model buzzed to be using RapidIO 10xN technology accelerated to 50 Gbps.

In addition, there is always a scope of innovation linked with the technology, where the platform can form its power with appropriate blend of processor types and accelerator capacity to optimize workloads at the edge of the wireless network. The modular building blocks can be installed by carriers across wireless base station, Cloud Radio Access Network (C-RAN) as well as in the central office for further server operations.

For greater capabilities five 1U systems can be formed in a mesh-network that can ensure up to 60 teraflops of computing capacity. Whereas, it takes 40 1U systems to be embedded together, in order to set up an industry standard data center rack to deliver up to half a petaflop of computing performance, in any central office.