Energy Aware Service Matchmaking in IoT Systems
碩士 === 國立臺灣大學 === 資訊工程學研究所 === 102 === The world is seeing more sensing and actuating devices being deployed in our environment. One issue for perpetually running Internet of Things (IoT) services is the energy efficiency. Many new IoT devices are running on more powerful platforms that have suffic...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Published: |
2014
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/40358071250013735626 |
| Summary: | 碩士 === 國立臺灣大學 === 資訊工程學研究所 === 102 === The world is seeing more sensing and actuating devices being deployed in our environment. One issue for perpetually running Internet of Things (IoT) services is the energy efficiency. Many new IoT devices are running on more powerful platforms that have sufficient computing and memory capabilities to support multiple sensors. Therefore, one energy saving strategy is to co-locate several energy-hungry services on one device in order to reduce the communication energy cost. This thesis proposes an energy aware service matchmaking approach for deploying an application onto IoT devices. The energy aware service matchmaking problem can be modeled as a Maximum Weighted Independent Set (MWIS) problem and we prove it to be a NP-hard problem. We design the algorithm to transform a service flow to a co-location graph, and then use heuristic algorithms to find the maximum independent set from the graph which will be used for making deployment decisions. The time complexity of transformation algorithm is studied. The performance of different co-locating algorithms are evaluated by simulation in our study. Simulation result shows that our proposed methods can save more 10\% energy than baseline.
|
|---|