U Illinois Research Reduces Serial Link Power Consumption

Researchers from the University of Illinois at Urbana-Champaign have designed technology that can help reduce the power consumption of microprocessor serial links in data centers.

According to information from Semiconductor Research Corp. (SRC), "serial links consume about 20 percent of microprocessor power and constitute about 7 percent of overall energy consumption in a data center," even though they are typically idle about 50 to 70 percent of the time. "These serial links are only sporadically used, such as when there is a request to access a Web page or a miss in the last level of cache."

The researchers designed a 7-gigabit per second (Gbps) transceiver that uses an on/off embedded clock architecture to significantly reduce the power consumption of microprocessor serial links. Previous technology reported power-on lock times of a couple of hundred nanoseconds in the case of memory interfaces to a few microseconds in the case of Energy Efficient Ethernet, but the transceiver technology designed by the U Illinois researchers reduced power-on lock times to less than 20 nanoseconds. They demonstrated the technology on 90 and 65 nanometer nodes, but according to SRC, it can also apply to memory and networking applications.

The researchers estimate that North American data centers could save $870 a year by using this transceiver technology.

SRC sponsored the research through the Texas Analog Center of Excellence (TxACE), and the team from the University of Illinois presented their research at the International Solid-State Circuits Conference (ISSCC) held in San Francisco, CA in February 2015.

Further information about the research can be found on the IEEE's site.

About the Author

Leila Meyer is a technology writer based in British Columbia. She can be reached at [email protected].

Featured

  • A Comprehensive Guide to the Best Value Evaluation Systems

    Choosing the most cost-effective evaluation system requires balancing price, usability and insight quality. In a landscape full of digital tools and data demands, it is important to prioritize platforms that deliver clear results without complicating operations.

  • handshake between two individuals with AI icons (brain, chip, network, robot) in the background

    Microsoft, Amazon Announce New Commitments to Support Presidential AI Challenge

    At the Sept. 4 meeting of the White House Task Force on Artificial Intelligence Education, Microsoft and Amazon unveiled new commitments to expanding AI education and skills training.

  • digital book with circuit patterns

    Turnitin and ACUE Partner on AI Training for Educators

    Turnitin is teaming up with the Association of College and University Educators to create a series of courses on AI and academic integrity designed to help faculty navigate the responsible use of AI in learning and assessment.

  • server racks, a human head with a microchip, data pipes, cloud storage, and analytical symbols

    OpenAI, Oracle Expand AI Infrastructure Partnership

    OpenAI and Oracle have announced they will develop an additional 4.5 gigawatts of data center capacity, expanding their artificial intelligence infrastructure partnership as part of the Stargate Project, a joint venture among OpenAI, Oracle, and Japan's SoftBank Group that aims to deploy 10 gigawatts of computing capacity over four years.