EU project boasts 100x energy efficiency in edge computing

EU project boasts 100x energy efficiency in edge computing
Technology News |
The LEGaTO project has produced a low power heterogeneous cloud-to-edge computing platform, spinning out a full stack RISC-V processor ecosystem.
By Jean-Pierre Joosting

Share:

The growth of the Internet of Things, edge and fog computing and the drive to reduce power consumption is putting significant pressure on hardware and software designers. Started to address this isseu, the three year European LEGaTO project has seen 100x increase in energy efficiency as well as gains on fault-tolerance, security and programmability.

The technologies developed in the project provide up to two orders of magnitude energy savings for five widely applicable use cases, form the IoT to smart homes. The project results will be used in different levels of European academia, research and industry, with EmbeDL as a spin-off company and the development of a RISC-V full stack.

“The work carried out in LEGaTO has already had a significant impact, improving the  products and services of our industrial partners and influencing a large number of European research projects, relevant standardization bodies and diverse academic programmes” said Osman Unsal, group manager for the Department of Computer Architecture for Parallel Paradigms at the Barcelona Supercomputing Center (BSC) and coordinator of LEGaTO.

A Cloud to Edge Microserver Platform covers most of the domains from cloud, edge and embedded computing by using a microserver architecture and a scalable system. It includes the LEGaTO Edge Server, which was completely designed and built in the project and focuses on edge and embedded applications.

All the components used and developed in the project are arranged in the LEGaTO stack, which gives an overview of the system developed in the project, from use cases to programming model, compiler and high-level synthesis languages, runtime, middleware and hardware.

 

Next: LEGaTO project spinouts


Based on the energy-efficient hardware platform and the task-based programming environment developed, five LEGaTO use cases were significantly optimised.

  • The use of FPGAs resulted in a 822x speedup in the biomarker discovery use case, which enables faster biomarker analysis.
  • The use of shared-memory programming style on distributed GPUs led to 10x energy savings in the smart home use case, the SmartMirror.
  • The smart city use case on operational urban pollutant dispersion modelling had a 7x gain in energy efficiency thanks to the use of FPGAs.
  • Up to 16x gain in energy efficiency and performance was achieved in the machine learning use case, devoted to automated driving and graphics rendering, using the EmbeDl optimizer. This use case led to the creation of EmbeDL, which improves energy efficiency and execution time in deep learning (DL) inference optimization tools.
  • The Secure IoT Gateway was vital to simplify the complexity of communication of local devices to a network, and it supported the above mentioned use cases to achieve their goals by reducing the complexity of security.
  • The various elements of the software toolset have been integrated together to provide a unified look and feel that facilitates the porting of future use cases to the energy-efficient LEGaTO hardware/software platform.

Three upcoming European projects will continue the development of the results achieved in the LEGaTO project. LV-EmbeDL is a partnership between BSC and EmbeDL funded by Tetramax that will implement and demonstrate the FPGA undervoltage techniques developed in LEGaTO.

The eProcessor project, coordinated by BSC and with other three LEGaTO partners in the consortium, will deliver the first completely open source European full stack ecosystem based on RISC-V technology.

The VEDLIoT (Very Efficient Deep Learning in IoT) project brings together 5 LEGaTO partners and is led by Bielefeld University to develop an IoT platform that uses deep learning algorithms distributed throughout the IoT continuum to achieve higher performance and energy efficiency.

www.legato-project.eu
www.EmbeDL.ai

 

Linked Articles
eeNews Wireless
10s