Latency aware deployment in the edge-cloud environment
Thesis title in Czech: | Nasazení aplikací zohledňující komunikační zpoždění v prostředí tzv. edge-cloud |
---|---|
Thesis title in English: | Latency aware deployment in the edge-cloud environment |
Key words: | edge-cloud, výkon, predikce, klastrová-analýza, regresní-analýza |
English key words: | edge-cloud, performance, prediction, cluster-analysis, regression-analysis |
Academic year of topic announcement: | 2018/2019 |
Thesis type: | diploma thesis |
Thesis language: | angličtina |
Department: | Department of Distributed and Dependable Systems (32-KDSS) |
Supervisor: | doc. RNDr. Petr Hnětynka, Ph.D. |
Author: | hidden - assigned and confirmed by the Study Dept. |
Date of registration: | 08.02.2019 |
Date of assignment: | 11.02.2019 |
Confirmed by Study dept. on: | 15.02.2019 |
Date and time of defence: | 06.02.2020 09:00 |
Date of electronic submission: | 06.01.2020 |
Date of submission of printed version: | 06.01.2020 |
Date of proceeded defence: | 06.02.2020 |
Opponents: | RNDr. David Bednárek, Ph.D. |
Advisors: | doc. RNDr. Iveta Hnětynková, Ph.D. |
Guidelines |
Edge-cloud computing is a new computing paradigm, which decentralizes computing to nodes that are geographically closer to client (compared the the traditional centralized cloud environments). Thanks to this decentralization, it is possible to achieve smaller communication latencies. However, for latency-sensitive applications (like augmented reality or coordinated movement of autonomous devices) the dislocation to edge-nodes is not enough as the end-to-end response is heavily influenced by the computation on the nodes. Thus in addition to cutting the communication latencies, it is necessary to provide guarantees on computation time for the workload process in the edge-cloud.
This calls for an additional layer on top of the edge-cloud, which manages deployment and redeployment of services in edge-cloud based on specification of timing requirements of individual services running the cloud. The goal of this thesis is to propose such a layer, which, given the knowledge of services' performance and resource consumption, predicts service's response time and finds optimal deployment (with regards to the application's real-time requirements on communication latencies). |
References |
Hnětynka P., Kubát P., Al Ali R., Gerostathopoulos I., Khalyeyev D.: Guaranteed Latency Applications in Edge-Cloud Environment, In ECSA 2018 Companion proceedings, Madrid, Spain, 2018
Östberg, P. O., J. Byrne, P. Casari, P. Eardley, A. F. Anta, J. Forsman, J. Kennedy, et al. “Reliable Capacity Provisioning for Distributed Cloud/Edge/Fog Computing Applications.” In 2017 European Conference on Networks and Communications (EuCNC), 1–6, 2017. Wang, S., M. Zafer, and K. K. Leung. “Online Placement of Multi-Component Applications in Edge Computing Environments.” IEEE Access 5 (2017): 2514–33. Oueis, J., E. C. Strinati, and S. Barbarossa. “The Fog Balancing: Load Distribution for Small Cell Cloud Computing.” In 2015 IEEE 81st Vehicular Technology Conference (VTC Spring), 1–6, 2015. |