Skip to main content

Advertisement

Fig. 8 | EURASIP Journal on Wireless Communications and Networking

Fig. 8

From: DIM: a distributed air index based on MapReduce for spatial query processing in road networks

Fig. 8

Tuning time access latency of four index approaches for kNN queries versus object density. a Tuning time (OL). b Tuning time (SR). c Tuning time (CAL). d Tuning time FL. e Access latency (OL). f Access latency (SR). g Access latency (CAL). h Access latency (FL). It can be seen that the adaptation of ISW is poor when the object density increases; at the same time, both the access latency and tuning time have increasing trend. The DIM, NPI, and IEI are relatively stable, especially the DIM index proposed in this paper; its tuning time and access latency are only one third of the NPI method and one half of the IEI method, especially in the large dataset FL. The IEI and NPI methods have a certain range of fluctuations, while the DIM method remains well stable. The object density increases, resulting in the expansion of the broadcast cycle, so the tuning time of ISW index increases rapidly. However, DIM method can quickly position the needed data according to the primary index or frame index parameters, without monitoring other frames and in doze mode, which greatly reduce the overhead

Back to article page