Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Evaluating Network Performance of Containerized Test Framework for Distributed Space Systems

Vaughan, Walter B. (2022) Evaluating Network Performance of Containerized Test Framework for Distributed Space Systems. Master's Thesis, University of Pittsburgh. (Unpublished)

[img]
Preview
PDF
Updated Version

Download (656kB) | Preview

Abstract

Distributed space systems are a mission architecture consisting of multiple spacecraft as a cohesive system in an attempt to provide multipoint sampling, increased mission coverage, or improved sample resolution, while reducing mission risk by through redundancy. For these reasons, distributed space systems are an increasingly common architecture for scientific space missions and broader space-related mission concepts. To fully realize the potential of these systems, eventually scaling to large-scale missions with hundreds or thousands of spacecraft, distributed space systems need to be operated as a single entity, which will enable a variety of novel scientific space missions.
By leveraging software containers, a technology developed for scalable terrestrial computing, distributed spacecraft systems can be simulated at scale entirely in software. Containers can run flight software in isolation, with multiple containers operating concurrently as a facsimile for in-situ spacecraft swarm behavior. This containerization approach allows developers to identify emergent behavior early and build confidence in the performability of the system, while limiting development overhead associated with higher system complexity.
For this container-based approach to simulation to be effective, the demonstrated system behavior must not be artificially influenced by the framework itself. To test this, a series of experiments are performed to quantify the available virtual network bandwidth on a single-machine host as a function of spacecraft swarm size, examining up to 32 fully connected containers. These tests invoke and measure communications throughout a network of containers in bidirectional network configurations, approximating the upper limit of possible network strain in such a system.
The container networks tested were able to simultaneously communicate at over 10 Gbps bidirectionally for up to 4 containers and around 1 Gbps bidirectionally for up to 16 containers. When containers communicated at fixed bit rates, no packet loss was observed at 500 kbps in any tested network size. With up to 32 fully connected containers at 10 Mbps, no more than 1.5% loss was recorded, with most connections reporting no packet loss. These results help clarify the expected fidelity of spacecraft simulation using virtual networks of containerized spacecraft software and establish expected performance limits for other containerized flight software development.


Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Vaughan, Walter B.wbv1@pitt.eduwbv10000-0002-5140-3526
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairGeorge, Alan D.alan.george@pitt.eduadg910000-0001-9665-2879
Committee MemberHu, Jingtongjthu@pitt.edujthu0000-0003-4029-4034
Committee MemberAbdelhakim, Maimaia@pitt.edumaia0000-0001-8442-0974
Date: 10 June 2022
Date Type: Publication
Defense Date: 5 April 2022
Approval Date: 10 June 2022
Submission Date: 29 March 2022
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 30
Institution: University of Pittsburgh
Schools and Programs: Swanson School of Engineering > Electrical and Computer Engineering
Degree: MS - Master of Science
Thesis Type: Master's Thesis
Refereed: Yes
Uncontrolled Keywords: Space,Spacecraft,Flight,Software,Container,Docker,Virtualization,Test,Swarm,Autonomy
Date Deposited: 10 Jun 2022 18:50
Last Modified: 10 Jun 2022 18:50
URI: http://d-scholarship.pitt.edu/id/eprint/42423

Metrics

Monthly Views for the past 3 years

Plum Analytics


Actions (login required)

View Item View Item