IDT launches low-latency computing platform

Integrated Device Technology has launched a heterogeneous mobile edge computing platform that performs real-time Big Data analytics and deep learning with low-latency computing at the network edge.

  • 9 years ago Posted in
Developed at IDT’s Open High-Performance Analytics and Computing (HPAC) lab, the platform utilizes modules currently in production and is based on innovations and collaboration with NVIDIA, Prodrive Technologies and Concurrent Technologies PLC.
 
 
“Network operators are looking for ways to add analytics capabilities on large amounts of unstructured data close to users, by pushing more compute capacity to the network edge, rather than sending  data on the backhaul to the core of the network data centers,” said Sailesh Chittipeddi, chief technology officer and vice president of Global Operations at IDT. “IDT’s Mobile Edge Computing Platform offers network and data center operators, as well as systems integrators, a quick path to computing at the edge by connecting accelerators based on a heterogeneous architecture with IDT’s low-latency RapidIO products.
 
“This is becoming more important with high-bandwidth streaming video and associated analytics that need to happen close to the last mile where users are located,” Chittipeddi noted.
 
The new platform is based on a 1U form factor server and features heterogeneous computing with NVIDIA® Tegra® K1 processors and x86 CPUs. The computing elements are connected using RapidIO fabric technology with about 100ns switching latency and 20 Gbps ports, with upcoming versions using RapidIO 10xN technology accelerated to 50 Gbps. The platform can be populated with appropriate mixes of processor types and accelerator capacity to optimize workloads for deep learning and analytics at the edge of the wireless network. Co-located with base stations, it is possible to support in ATCA form factors if required. The modular building blocks can be deployed by carriers for server functionality at the wireless base station, Cloud Radio Access Network (C-RAN) or in the central office. And with distributed data center functions becoming increasingly important for streaming video applications, the platform can also be deployed at the network point of presence.  
 
Five 1U systems can be meshed together in a network that can support up to 60 teraflops of computing capacity using the energy-efficient NVIDIA Tegra processors. In a central office, up to 40 1U systems can be connected in an industry-standard data center rack to deliver up to half a petaflop of compute performance.
 
"Prodrive Technologies has developed an extensive suite of RapidIO-based server and supercomputing solutions for a variety of markets,” said Pieter Janssen CEO of Prodrive Technologies. "Our latest collaboration with IDT on heterogeneous computing is in the initial Mobile Edge Computing Platform, which builds on our RapidIO top-of-rack switching solutions already developed for both C-RAN and high-performance computing markets."
The 2024 State of Data Intelligence Report finds companies struggling with AI governance more than...
On average, only 48% of digital initiatives meet or exceed business outcome targets, according to...
Fivetran equips over half of Trinny London's workforce with self-service analytics, accelerating...
Techcombank, one of Vietnam’s leading financial institutions, has implemented the Databricks Data...
New survey data from Cohesity reveals that consumers surveyed worldwide are highly concerned about...
As the speed of decisions increases, new Confluent research shows half of C-level executives are...
NinjaOne AI program focuses on customer success and thoughtful adoption over hype.
The new seven-story Fitzrovia-based space will be one of the company's largest offices outside of...