Scheduling Multi-Component Applications Across Federated Edge Clusters With Phare
The shift towards agile microservice architecture has enabled significant benefits for IT companies but has also resulted in increased complexity for Cloud orchestration tools. Traditional tools were designed for centralized data centers and are ineffective for locating microservices in geographical...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Open Journal of the Communications Society |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10473112/ |
| _version_ | 1797227279322972160 |
|---|---|
| author | Gabriele Castellano Stefano Galantino Fulvio Risso Antonio Manzalini |
| author_facet | Gabriele Castellano Stefano Galantino Fulvio Risso Antonio Manzalini |
| author_sort | Gabriele Castellano |
| collection | DOAJ |
| description | The shift towards agile microservice architecture has enabled significant benefits for IT companies but has also resulted in increased complexity for Cloud orchestration tools. Traditional tools were designed for centralized data centers and are ineffective for locating microservices in geographically-distributed edge-like infrastructures. This paper presents Phare, a decentralized scheduling algorithm designed to optimize the placement of microservices by satisfying their computing and communication demands while minimizing deployment costs. Phare employs a heuristic-based approach to solve the NP-Hard scheduling problem, prioritizing the microservices with the more stringent requirements and placing them on the most convenient computing facilities, based on the concept of affinity, contributing to the field by providing a more holistic approach to resource scheduling in edge computing. We validate our approach against Firmament, the state-of-the-art workload scheduling algorithm for component-based applications, on simulated edge infrastructures with hundreds of clusters. Phare achieves up to a <inline-formula> <tex-math notation="LaTeX">$10\times $ </tex-math></inline-formula> reduction in terms of deployment costs compared to Firmament while providing a much lower scheduling latency. |
| first_indexed | 2024-04-24T14:38:17Z |
| format | Article |
| id | doaj.art-2e06441e17cf47cf8dc015f24a011441 |
| institution | Directory Open Access Journal |
| issn | 2644-125X |
| language | English |
| last_indexed | 2024-04-24T14:38:17Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of the Communications Society |
| spelling | doaj.art-2e06441e17cf47cf8dc015f24a0114412024-04-02T23:00:54ZengIEEEIEEE Open Journal of the Communications Society2644-125X2024-01-0151814182610.1109/OJCOMS.2024.337791710473112Scheduling Multi-Component Applications Across Federated Edge Clusters With PhareGabriele Castellano0https://orcid.org/0000-0002-1889-7675Stefano Galantino1https://orcid.org/0000-0002-3386-6471Fulvio Risso2https://orcid.org/0000-0001-6134-7890Antonio Manzalini3https://orcid.org/0000-0003-1633-3099Centre Inria d’Université Côte d’Azur, Valbonne, FranceDepartment of Computer and Control Engineering, Politecnico di Torino, Turin, ItalyDepartment of Computer and Control Engineering, Politecnico di Torino, Turin, ItalyInnovation Labs, Telecom Italia Mobile, Turin, ItalyThe shift towards agile microservice architecture has enabled significant benefits for IT companies but has also resulted in increased complexity for Cloud orchestration tools. Traditional tools were designed for centralized data centers and are ineffective for locating microservices in geographically-distributed edge-like infrastructures. This paper presents Phare, a decentralized scheduling algorithm designed to optimize the placement of microservices by satisfying their computing and communication demands while minimizing deployment costs. Phare employs a heuristic-based approach to solve the NP-Hard scheduling problem, prioritizing the microservices with the more stringent requirements and placing them on the most convenient computing facilities, based on the concept of affinity, contributing to the field by providing a more holistic approach to resource scheduling in edge computing. We validate our approach against Firmament, the state-of-the-art workload scheduling algorithm for component-based applications, on simulated edge infrastructures with hundreds of clusters. Phare achieves up to a <inline-formula> <tex-math notation="LaTeX">$10\times $ </tex-math></inline-formula> reduction in terms of deployment costs compared to Firmament while providing a much lower scheduling latency.https://ieeexplore.ieee.org/document/10473112/Resource sharingcloud-to-edgeservice allocation |
| spellingShingle | Gabriele Castellano Stefano Galantino Fulvio Risso Antonio Manzalini Scheduling Multi-Component Applications Across Federated Edge Clusters With Phare IEEE Open Journal of the Communications Society Resource sharing cloud-to-edge service allocation |
| title | Scheduling Multi-Component Applications Across Federated Edge Clusters With Phare |
| title_full | Scheduling Multi-Component Applications Across Federated Edge Clusters With Phare |
| title_fullStr | Scheduling Multi-Component Applications Across Federated Edge Clusters With Phare |
| title_full_unstemmed | Scheduling Multi-Component Applications Across Federated Edge Clusters With Phare |
| title_short | Scheduling Multi-Component Applications Across Federated Edge Clusters With Phare |
| title_sort | scheduling multi component applications across federated edge clusters with phare |
| topic | Resource sharing cloud-to-edge service allocation |
| url | https://ieeexplore.ieee.org/document/10473112/ |
| work_keys_str_mv | AT gabrielecastellano schedulingmulticomponentapplicationsacrossfederatededgeclusterswithphare AT stefanogalantino schedulingmulticomponentapplicationsacrossfederatededgeclusterswithphare AT fulviorisso schedulingmulticomponentapplicationsacrossfederatededgeclusterswithphare AT antoniomanzalini schedulingmulticomponentapplicationsacrossfederatededgeclusterswithphare |