Reducing Latency in Cloud-Based Fuel Monitoring Systems
Fuente:
Zenodo
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Recurso digital |
| Pubblicazione: |
Zenodo
2021
|
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866902266601013248 |
|---|---|
| author | Rohith Varma Vegesna |
| author_facet | Rohith Varma Vegesna |
| contents | <p>Cloud-based fuel monitoring systems are increasingly adopted to provide continuous oversight of station operations, harnessing automated data collection, rapid analytics, and near-real-time decision-making. These systems handle substantial data from Automatic Tank Gauge (ATG) devices and pump dispensers, making low latency crucial for accurate inventory tracking, timely leak detection, and rapid response to operational anomalies. Yet, designing an end-to-end pipeline that consistently achieves minimal latency, particularly under varying network conditions, poses considerable technical challenges.<br>This paper introduces an integrated approach that capitalizes on managed services within a well-established public cloud environment to lower round-trip times and processing delays. The proposed architecture utilizes edge processing for buffering and compression, reducing data volume and mitigating connection disruptions. The ingestion pipeline relies on fully managed, event-driven services to orchestrate serverless functions. These functions each serve as dedicated microservices, handling tasks such as anomaly detection, inventory reconciliation, and real-time alerts. Processed data is stored in low-latency databases for immediate access through customized dashboards, offering an operational view that updates within a matter of seconds.<br>A pilot study demonstrates the architecture’s effectiveness, highlighting measurable gains in responsiveness and uptime when compared to legacy systems. Substantial reductions in latencycoupled with improved operational efficiencyunderscore the benefits of separating critical tasks into discrete microservices, leveraging agile deployment practices, and incorporating automated monitoring. This work concludes with a discussion of potential enhancements, including scaling to diverse station networks, integrating predictive analytics, and reinforcing security at both the edge and the cloud layers.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_14905652 |
| institution | Zenodo |
| language | |
| publishDate | 2021 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Reducing Latency in Cloud-Based Fuel Monitoring Systems Rohith Varma Vegesna <p>Cloud-based fuel monitoring systems are increasingly adopted to provide continuous oversight of station operations, harnessing automated data collection, rapid analytics, and near-real-time decision-making. These systems handle substantial data from Automatic Tank Gauge (ATG) devices and pump dispensers, making low latency crucial for accurate inventory tracking, timely leak detection, and rapid response to operational anomalies. Yet, designing an end-to-end pipeline that consistently achieves minimal latency, particularly under varying network conditions, poses considerable technical challenges.<br>This paper introduces an integrated approach that capitalizes on managed services within a well-established public cloud environment to lower round-trip times and processing delays. The proposed architecture utilizes edge processing for buffering and compression, reducing data volume and mitigating connection disruptions. The ingestion pipeline relies on fully managed, event-driven services to orchestrate serverless functions. These functions each serve as dedicated microservices, handling tasks such as anomaly detection, inventory reconciliation, and real-time alerts. Processed data is stored in low-latency databases for immediate access through customized dashboards, offering an operational view that updates within a matter of seconds.<br>A pilot study demonstrates the architecture’s effectiveness, highlighting measurable gains in responsiveness and uptime when compared to legacy systems. Substantial reductions in latencycoupled with improved operational efficiencyunderscore the benefits of separating critical tasks into discrete microservices, leveraging agile deployment practices, and incorporating automated monitoring. This work concludes with a discussion of potential enhancements, including scaling to diverse station networks, integrating predictive analytics, and reinforcing security at both the edge and the cloud layers.</p> |
| title | Reducing Latency in Cloud-Based Fuel Monitoring Systems |
| url | https://doi.org/10.5281/zenodo.14905652 |