A study on network performance metrics and their composition

Abstract

Abstract:Research backbone networks like GÉANT2 and the National Research and Education Networks are used by a variety of scientists and research projects. These users and the network engineers operating the networks would like to get access to network performance metrics to optimise their use of the network and to troubleshoot performance degradations, when they happen. A variety of tools for performing network measurements already exist, and the perfSONAR architecture developed within the Joint Research Activity 1 (JRA1) of GÉANT2 aims at integrating them in a coherent framework. However, a harmonised definition of which metrics are mostly interesting and how measurements must be carried out is still lacking. In this paper we suggest the set of elementary metrics which are more relevant, along with indication about how to post process (or "transform", or "compose") them in order to obtain derived summary values that can quickly and intuitively give an indication of network performance. Methods to perform the composition are presented, together with constraints which have to be taken into account to get accurate results. In particular, delay measurements are the most delicate ones to compose. We carried out a series of experiments for proofing the validity of composition of delay metrics, and we briefly present some preliminary results.

Proactive & Time-Optimized Data Synopsis Management at the Edge

“…Each EN apart from maintaining synopsis, monitors, nowcasts and forecasts network performance metrics to assess the optimal synopsis delivery time. Network performance metrics are categorized into [25]: (i) communication availability; (ii) packet loss and error; (iii) end-to-end latency; (iv) throughput (volume of data transferred per time unit). There are many efforts for monitoring packet error loss, latency and throughput while operating systems offer access to such parameters (e.g., via files -/proc/net/dev or commands -netperf in Linux).…”

“…8 shows pd f (t ⇤ ) densities having a 'bad' performance network with E[b ] ⇡ E[b ] ⇡ 0.70. Here, OSDM's update times range in (1,10) and (1,25) in D1 and D2, resp. This confirms our observations related to the intention of OSDM to accumulate update quanta while postponing the decision due to bad network performance.…”

An efficient scheme for applying software updates in pervasive computing applications

“…In our model, we focus on two types of metrics related to (i) the network , and, (ii) the performance of a node. Network performance metrics can be categorized into [22]: (i) availability; (ii) packet loss and error ; (iii) delay;…”

New Innovations in Performance Analysis of Computer Networks: A Review

“…Performance metrics are the criteria used to evaluate the performance of a system [7]. There are a number of metrics applicable to computer networks, but the focus of this write-up will be on those metrics which are of greatest relevance to network performance [9]. New innovations are being employed to accurately monitor, measure and study their effects on network performance.…”