Table of Contents
Fetching ...

An experimental study of the response time in an edge-cloud continuum with ClusterLink

Marc Michalke, Fin Gentzen, Admela Jukan, Kfir Toledo, Etai Lev Ran

TL;DR

This study investigates how inter-cluster communication in an IoT-Edge-Cloud continuum affects application response times using the open-source ClusterLink solution. The authors build an experimental testbed with Nebula overlay networking and two Kubernetes clusters running Knative serverless functions, comparing ClusterLink performance against direct endpoint communication across local, edge, and cloud network scenarios. Results show that ClusterLink imposes negligible overhead for small payloads after initial session setup, but larger payloads experience latency increases that scale with the underlying inter-cluster link delay, reflecting TCP and encryption processing effects and the current early-stage status of the software. The work provides practical guidance for deploying inter-cluster communication in edge environments and identifies areas for future work, including broader hardware testing, removal of overlay overhead, and benchmarking against established inter-cluster networks.

Abstract

In this paper, we conduct an experimental study to provide a general sense of the application response time implications that inter-cluster communication experiences at the edge at the example of a specific IoT-edge-cloud contiuum solution from the EU Project ICOS called ClusterLink. We create an environment to emulate different networking topologies that include multiple cloud or edge sites scenarios, and conduct a set of tests to compare the application response times via ClusterLink to direct communications in relation to node distances and request/response payload size. Our results show that, in an edge context, ClusterLink does not introduce a significant processing overhead to the communication for small payloads as compared to cloud. For higher payloads and on comparably more aged consumer hardware, ClusterLink version 0.2 introduces communication overhead relative to the delay experienced on the link.

An experimental study of the response time in an edge-cloud continuum with ClusterLink

TL;DR

This study investigates how inter-cluster communication in an IoT-Edge-Cloud continuum affects application response times using the open-source ClusterLink solution. The authors build an experimental testbed with Nebula overlay networking and two Kubernetes clusters running Knative serverless functions, comparing ClusterLink performance against direct endpoint communication across local, edge, and cloud network scenarios. Results show that ClusterLink imposes negligible overhead for small payloads after initial session setup, but larger payloads experience latency increases that scale with the underlying inter-cluster link delay, reflecting TCP and encryption processing effects and the current early-stage status of the software. The work provides practical guidance for deploying inter-cluster communication in edge environments and identifies areas for future work, including broader hardware testing, removal of overlay overhead, and benchmarking against established inter-cluster networks.

Abstract

In this paper, we conduct an experimental study to provide a general sense of the application response time implications that inter-cluster communication experiences at the edge at the example of a specific IoT-edge-cloud contiuum solution from the EU Project ICOS called ClusterLink. We create an environment to emulate different networking topologies that include multiple cloud or edge sites scenarios, and conduct a set of tests to compare the application response times via ClusterLink to direct communications in relation to node distances and request/response payload size. Our results show that, in an edge context, ClusterLink does not introduce a significant processing overhead to the communication for small payloads as compared to cloud. For higher payloads and on comparably more aged consumer hardware, ClusterLink version 0.2 introduces communication overhead relative to the delay experienced on the link.
Paper Structure (10 sections, 4 figures, 3 tables)

This paper contains 10 sections, 4 figures, 3 tables.

Figures (4)

  • Figure 1: Software Stack and Infrastructure
  • Figure 2: Network Scenarios
  • Figure 3: Hello-World Response Time
  • Figure 4: Payload 100KB