Light Reading

Fujitsu Labs Develops 56G Receiver Circuit for Server Comms

Light Reading
News Wire Feed
Light Reading
6/13/2014
50%
50%

KAWASAKI, Japan -- Fujitsu Laboratories Ltd. today announced the development of a receiver circuit capable of receiving communications at 56 Gbps. This marks the world's fastest data communications between CPUs equipped in next-generation servers.

In recent years, raising data-processing speeds in servers has meant increasing CPU performance, together with boosting the speed of data communications between chips, such as CPUs. However, one obstacle to this has been improving the performance of the circuits that correct degraded waveforms in incoming signals.

Fujitsu Laboratories has used a new "look-ahead" architecture in the circuit that compensates for quality degradation in incoming signals, parallelizing the processing and increasing the operating frequency for the circuit in order to double its speed.

This technology holds the promise of increasing the performance of next-generation servers and supercomputers.

Details of this technology are being presented at the 2014 Symposia on VLSI Technology and Circuits, opening June 9 in Hawaii (VLSI Circuits Presentation 11-2).

In order to enhance the performance of datacenters underpinning the spread of cloud computing in recent years, a need has arisen for servers that process data faster. While this can be achieved partly through faster CPUs, large-scale systems connecting many CPUs are also being built, and the amount of data transmitted, either within the same CPU-equipped chassis or across separate chassis, is growing dramatically. To cope with these volumes, data communication speeds in the current generation of servers is increasing from a few gigabits per second today to ten or more gigabits per second. Because it is anticipated that data processing volumes will continue to experience explosive growth, however, for the next generation of high-performance servers, the goal is to double current levels to 56 Gbps. Furthermore, the Optical Internetworking Forum (OIF) is moving forward on the standardization of 56 Gbps for the optical modules used for optical transmission between chassis.

An effective way to speed up the receiver circuit is to improve the processing performance of the decision feedback equalizer (DFE) circuit that compensates for the degraded input-signal waveform (Figure 2).The principle behind DFE is to correct the input signal based on the bit-value of the previous bit and to emphasize changes in the input signal, but the actual circuit design works by choosing between two predefined corrected candidates. If the previous bit value was a 0, the correction process would apply a positive correction to the input signal (additive) to emphasize the change from 0 to 1. If the previous bit value was 1, it would apply a negative correction to the input signal (subtractive) to emphasize the change from 1 to 0. If another 0 was received, the positive compensation would increase the signal level, but not to such a level as would create a problem for the 1/0 decision circuit.

In ordinary circuit designs that run at 56 Gbps, there are 16 DFE circuits coupled together. Using 4 DFE circuits as an example, they run at 1/4th the actual frequency. So for 28-Gbps communications rates, 1/4th of that is 142 picoseconds, and four bits-worth of compensation can be applied during that interval. But at 56 Gbps, 1/4th of that speed amounts to 71 picoseconds, during which time only 2 bits-worth of compensation can be applied, resulting in timing errors (Figure 3).

Fujitsu Laboratories took a new approach, a "look-ahead" method that can be implemented as a parallel process, pre-calculating two candidates based on the selection result for the previous bit, and simultaneously deciding the value of the previous bit and the current bit after deciding the value of the bit two bits previous. This shortens calculation times, resulting in a receiver circuit that can operate at 56 Gbps (Figure 4).

Features of the new technology are as follows:

1. Look-ahead compensation process

In the existing method, the result of the previous bit's selection circuit (A) is implemented by a circuit combining the result of the selection circuit for the bit two bits previous (B) and the input signal for the selection circuit one bit previous (+/- compensation data) (C). In the look-ahead method, the input signal for the selection circuit one bit previous (+/- compensation data) (D) and the input signal for the selection circuit of the current bit (+/- compensation data) (E) are combined using a look-ahead circuit, and candidates for the selection circuit are pre-computed. Doing this relies on only the result from the selection circuit for the bit two bits previous, without using the result from the selection circuit for the bit one bit previous, while functioning essentially the same as the existing method.

2. Parallelized look-ahead processing using a hold circuit

Multiple look-ahead circuits that apply DFE one bit at a time can operate independently (Figure 5). Fujitsu Laboratories inserted a hold circuit between the selection circuit and look-ahead circuit, with the input and output of each hold circuit being synchronized, making it possible to parallelize these processes.

Because the calculation time for the look-ahead circuit is roughly the same as the selection time for the selector, overall calculation time is dependent on the number of selectors deciding based on data from two bits previous, so in a four-bit structure, that would be two. Running at 1/4th of 56 Gbps allows computations to be safely completed within 71 picoseconds. This makes it possible to receive data at 56 Gbps, doubling existing communications speeds.

This technology makes it possible to increase bandwidth of communications between CPUs in future servers and supercomputers, even if CPU performance doubles, without increasing pin counts, and will contribute to increased performance in large-scale systems where numerous CPUs are interconnected.

In addition, it complies with standards for optical-module communications, and compared to the 400-Gbps Ethernet in OIF-CEI-28G optical-module communications, the number of circuits running in parallel (number of lanes) can be halved, allowing for smaller optical modules running on less power, and higher system performance.

Fujitsu Laboratories plans to apply this technology to the interfaces of CPUs and optical modules, with the goal of a practical implementation in fiscal 2016. The company is also considering applications to next-generation servers, supercomputers, and other products.

Fujitsu Laboratories Ltd.

(0)  | 
Comment  | 
Print  | 
Newest First  |  Oldest First  |  Threaded View        ADD A COMMENT
Flash Poll
From The Founder
Network architects aiming to upgrade their networks to support agile, open, virtualized services in the 21st century need to consider new criteria when choosing between technology suppliers.
Live Streaming Video
CLOUD / MANAGED SERVICES: Prepping Ethernet for the Cloud
Moderator: Ray LeMaistre Panelists: Jeremy Bye, Leonard Sheahan
Wagner’s Ring
Business Value Trumps Moving Bits for Virtual Networks

7|7|15   |   01:22   |   (0) comments


At the recent Open Networking Summit, carriers described how they're focused on increasing revenue rather than just improving technology. This requires cultural changes, which can be more difficult than technology changes.
LRTV Custom TV
VeEX: Live from the Big Telecom Event 2015

7|7|15   |   2:28   |   (0) comments


VeEX's Carl Goldschmidt provides new product updates, including the RXT modular test platform and TX300s. With the latest technology in pluggable physical interfaces, the RXT platform combines multiple technologies with testing range from 10 Mbit/s to 100 Gbit/s. The TX300s is an "All-in-one" advanced multi-service test platform from 64 Kbit/s to 100 Gbit/s that ...
LRTV Custom TV
Light Reading Interviews MRV: 2015 Big Telecom Event

7|7|15   |   3:45   |   (0) comments


Scott St. John, SVP of Global Sales & Service, discusses MRV's customer-centric culture, the new OptiPacket OP-X1 100G access platform, and recent customer wins with Light Reading at BTE 2015.
LRTV Custom TV
Procera: Enhancing the Subscriber Experience

7|7|15   |   4:53   |   (0) comments


Hear from Procera on the latest in Subscriber Experience solutions. Learn how Procera's RAN Perspective technology helps network operators monitor and manage network traffic, in real-time, solving congestion and other network issues.
Between the CEOs
HP's Prodip Sen: The NFV Journey

7|7|15   |   21:39   |   (0) comments


Prodip Sen, CTO of the Network Functions Virtualization (NFV) business unit at HP, talks to Light Reading founder and CEO Steve Saunders about the four stages of the NFV journey.
LRTV Documentaries
Leading Lights 2015 Winners' Video

7|7|15   |   08:34   |   (2) comments


Hear what the Leading Lights 2015 winners had to say after they picked up their awards in Chicago.
LRTV Documentaries
PSA: The Perils of a Hyper-Connected Society

7|6|15   |   1:38   |   (0) comments


A public service announcement calling attention to the perils of a hyper-connected society.
LRTV Interviews
Guavus Takes Analytics on the Road

7|6|15   |   4:09   |   (3) comments


Guavus CEO Anukool Lakhina tells Light Reading CEO and founder Steve Saunders how the analytics tools developed for telcos are applicable to the transportation industry.
Telecom Innovators Video Showcase
Mellanox's New 25/50/100Gb/s Ethernet Products

7|6|15   |   2:44   |   (0) comments


Mellanox offers a complete 10/25/40/50/56/100Gbit/s solution that delivers industry-leading performance, scalability, reliability and efficiency for optimal application performance and data center ROI.
Telecom Innovators Video Showcase
Arista's CloudVision

7|6|15   |   6:01   |   (0) comments


Anshul Sadana answers questions from Steve Saunders, Light Reading’s founder and CEO, about Arista’s CloudVision, a global cloud network controller for workload orchestration and workflow automation delivering a turnkey solution for cloud networking.
LRTV Custom TV
Red Hat Demo

7|2|15   |   10:53   |   (0) comments


Red Hat's Nicolas Lemieux demonstrates how Red Hat is driving innovation through open source communities.
LRTV Custom TV
Red Hat's Approach to OpenStack Adoption

7|2|15   |   5:17   |   (0) comments


Red Hat's Radhesh Balakrishnan outlines his company's open source strategy for both enterprises and telcos.
Upcoming Live Events
September 16-17, 2015, The Westin Galleria Dallas, Dallas, TX
September 16, 2015, The Westin Galleria Dallas, Dallas, TX
September 29-30, 2015, The Westin Grand Müchen, Munich, Germany
October 6, 2015, The Westin Peachtree Plaza, Atlanta, GA
October 6, 2015, Westin Peachtree Plaza, Atlanta, GA
October 14-15, 2015, New Orleans Ernest N. Morial Convention Center, New Orleans, LA
November 5, 2015, Hilton Santa Clara, Santa Clara, CA
All Upcoming Live Events
Infographics
Equinix walks through the past four digital economy eras from computing to network to connected to today's interconneted world.
Hot Topics
What's a Gigabit Good For?
Carol Wilson, Editor-at-large, 7/1/2015
IBM Sows Seed for New Telco Unit, Plans NFV/SDN Push
Ray Le Maistre, Editor-in-chief, 7/6/2015
What the Helio? MVNO Is Back After First Flop
Sarah Thomas, Editorial Operations Director, 7/6/2015
FCC Chairman Talks Up SDN/NFV
Mari Silbey, Senior Editor, Cable/Video, 7/2/2015
A&E Launches Lifetime OTT Service
Alan Breznick, Cable/Video Practice Leader, 7/3/2015
Like Us on Facebook
Twitter Feed
Webinar Archive
BETWEEN THE CEOs - Executive Interviews
Prodip Sen, CTO of the Network Functions Virtualization (NFV) business unit at HP, talks to Light Reading founder and CEO Steve Saunders about the four stages of the NFV journey.
Casa Systems has been going from strength to strength over the last couple of years. In 2013, it became the first vendor to ship an integrated CCAP device -- the ...
Cats with Phones