Int. J. Cloud Computing, Vol. 9, No. 1, 2020 55
In-network processing for edge computing with
InLocus
Lucas R.B. Brasilino
Intelligent Systems Engineering – ISE,
School of Informatics, Computing and Engineering – SICE,
Indiana University,
Bloomington, IN, USA
Email: lbrasili@iu.edu
Naveen Marri
Computer Science – CS,
School of Informatics, Computing and Engineering – SICE,
Indiana University,
Bloomington, IN, USA
Email: navmarri@iu.edu
Alexander Shroyer, Catherine Pilachowski,
Ezra Kissel and Martin Swany*
Intelligent Systems Engineering – ISE,
School of Informatics, Computing and Engineering – SICE,
Indiana University,
Bloomington, IN, USA
Email: ashroyer@iu.edu
Email: calupila@iu.edu
Email: ezkissel@iu.edu
Email: swany@iu.edu
*Corresponding author
Abstract: As sensors and smart device infrastructure grows, networks
are increasingly heterogeneous and diverse. We propose an efficient and
low-latency architecture called InLocus, which facilitates stream processing
at the network’s edge. InLocus balances hardware-accelerated performance
with the flexibility of asynchronous software control. In this paper, we
extend InLocus architecture by implementing compute nodes in a more
traditional cloud-based solution in the form of Apache Kafka and Twitter
Heron framework, as well as by introducing a new runtime approach for
the previously handwritten C Server. We utilise a flexible platform (Xilinx
Zynq SoC) to compare microbenchmarks between the latter and high-level
synthesis (HLS) version in programmable hardware.
Keywords: in-network processing; edge computing; internet of things; IoT;
programmable logic; FPGA; offloading; hardware acceleration.
Copyright © 2020 Inderscience Enterprises Ltd.