Data structure enabling retrieval of time series of traffic with the requested granularity

Abstract

Some network management tasks require the fine-grained information of time series of traffic. However, it takes a large overhead to obtain the fine-grained information. One approach is to obtain the traffic information only with the required granularities because the traffic information with the fine granularity are not always necessary. In this approach, the fine-grained traffic information is stored at each monitoring point, and the network manager receive only the traffic information with the required granularity. The required granularity of information depends on time, but none of existing data structure enables immediate retrieval of time series of traffic with the required granularity that changes in time. Therefore, in this paper, we propose a data structure enabling it. Through the numerical evaluation, we demonstrate that our data structure enables immediate retrieval of time series of traffic without a large calculation time to store the traffic data.