.png){kind=icon}
.png){kind=icon}
Return
Multi-application operator placement in cloud-edge infrastructure for big data stream processing
.png){kind=icon}
PRE
AI
DOI：10.1007/s10586-026-06524-9.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Internet of Things (IoT) applications often require real-time processing of data streams generated by a rapidly growing number of geographically distributed smart devices. Traditionally, these data streams are sent to distant cloud servers for processing, leading to high end-to-end latency. To address this issue, recent studies have explored cloud-edge infrastructures, which combine cloud resources with edge devices closer to data sources. One of the key challenges in Data Stream Processing (DSP) within this hybrid environment is operator placement. While recent studies have focused on multi-objective operator allocation across heterogeneous cloud-edge devices, two major aspects remain unaddressed: 1) handling trade-offs between latency, network usage, product owner preferences, QoS constraints, and heterogeneous resource utilization and 2) managing multiple DSP applications running in parallel, rather than treating them sequentially. This paper proposes a resource-aware multi-application operator placement method that optimizes both end-to-end latency and network usage, while meeting QoS constraints and application owners’ preferences in heterogeneous cloud-edge environments. Furthermore, we develop a proof-of-concept prototype using Apache Storm. Experimental results on a real-world heterogeneous cloud-edge testbed demonstrate that the proposed method effectively manages the trade-off between latency and network usage.
Keywords:
Operator placement
Data stream processing
Multi-application
Multi-objective optimization
Cloud-edge computing
Journal
C
IF:
4.1
Papers:
5.0K
Citations:
7.5K