Characterizing the impact of topology on IoT stream processing
Characterizing the impact of topology on IoT stream processing
复制标题
DOI：
10.1109/wf-iot.2018.8355119
复制
发表时间：
2018-02
期刊：
影响因子：
--
通讯作者：
Anindya Dey;Kim Stuart;Matthew E. Tolentino
中科院分区：
文献类型：
--
作者：
Anindya Dey;Kim Stuart;Matthew E. Tolentino
物联网(IoT)通过将基于传感器的边缘设备链接到网络可访问的服务和资源来扩展传统的网络物理系统。在当前的大多数物联网部署中，传感器数据从边缘设备流向服务器进行存储。然后使用分析管道将这些原始传感器数据实时转换为可操作的信息。随着更多物联网设备的部署，服务器端接收的数据量和速率可能会大幅增加。这有可能抵消超出物联网分析系统可接受限制的响应延迟。在本文中，我们比较了不同的服务器端流处理拓扑对实时摄取和分析物联网传感器数据的影响。我们通过我们的实时物联网平台Namatad使用真实的建筑传感器数据。我们已经表征和分析了由于我们将数据传输到分析管道的接收和路由过程的不同粒度级别而导致的延迟和服务质量影响。我们的结果表明，随着物联网系统密度的不断扩展，针对物联网数据流的服务器端拓扑管理对于延迟敏感型控制和分析应用至关重要。
The Internet of Things (IoT) extends traditional cyber-physical systems by linking sensor based edge devices to network accessible services and resources. In most current IoT deployments, sensor data is streamed from edge devices to servers for storage. Analytical pipelines are then used to translate this raw sensor data into actionable information in real-time. As additional IoT devices are deployed, the volume and rate of data received on the server side can increase dramatically. This has a possibility of offsetting the response latencies beyond acceptable limits for IoT analytical systems. In this paper, we compare the impact of alternative serverside stream processing topologies for ingesting and analyzing IoT sensor data in real-time. We use real building sensor data with our real-time IoT platform called Namatad. We have characterized and analyzed the latency and QoS impact due to the different levels of granularity of the ingestion and routing process by which we transmit data into the analytical pipelines. Our results show that as IoT systems continue to scale in density, server-side topology management for IoT data streams is critical for latency-sensitive control and analysis applications.