Accurate and efficient SLA compliance monitoring

Abstract

Service level agreements (SLAs) define performance guarantees made by service providers, e.g, in terms of packet loss, delay, delay variation, and network availability. In this paper, we describe a new active measurement methodology to accurately monitor whether measured network path characteristics are in compliance with performance targets specified in SLAs. Specifically, (1) we describe a new methodology for estimating packet loss rate that significantly improves accuracy over existing approaches; (2) we introduce a new methodology for measuring mean delay along a path that improves accuracy over existing methodologies, and propose a method for obtaining confidence intervals on quantiles of the empirical delay distribution without making any assumption about the true distribution of delay; (3) we introduce a new methodology for measuring delay variation that is more robust than prior techniques; and (4) we extend existing work in network performance tomography to infer lower bounds on the quantiles of a distribution of performance measures along an unmeasured path given measurements from a subset of paths. We unify active measurements for these metrics in a discrete time-based tool called SLAM. The unified probe stream from SLAM consumes lower overall bandwidth than if individual streams are used to measure path properties. We demonstrate the accuracy and convergence properties of SLAM in a controlled laboratory environment using a range of background traffic scenarios and in one-and two-hop settings, and examine its accuracy improvements over existing standard techniques.

Uncovering Global Icebergs in Distributed Streams: Results and Implications

“…One example of an iceberg is a network traffic flow (e.g., all the traffic coming from a particular source IP address or targeting a particular destination IP address) that has high aggregate volume across many different monitors, even if it does not appear large at any single monitor. Detecting this type of event is important for a number of applications, including detecting DDoS attacks [1], finding heavy-hitters in Content Delivery Networks [2], discovering worms and other anomalies [3], as well as ensuring SLA compliance [4]. Applications that detect DDoS attacks need to find destination IP addresses that occur frequently across multiple ingress points.…”

Exploiting hybrid measurements for network troubleshooting

“…For instance, authors of [9] build an active measurement system to address the problem of spatially identifying faults in the network. Similarly, authors of [10] propose a complex multi-objective methodology based on a plethora of active measurements to specifically identify Service-level Agreement (SLA) violations. However, active measurement come at the cost of injecting artificial load into the network, and thus must be carefully designed to not overload the system.…”

Towards an Efficient Service Level Agreement Assessment

“…Some important work has been performed by Sommers et al in [3], where the authors present SLAm, an active probing tool, which implements innovative packet loss, delay, and delay variation estimation techniques for SLA assessment. In this research, the authors stress the need of proper metric estimation in order to lead to correct SLA assessment.…”

“…Both ISP and customers want to know at any time the quality of the network services, and whether it is respecting the contracted SLA. For this purpose, classical approaches for SLA assessment [1], [2], [3] focused on accurately measuring (when possible) or estimating (most of the times) network QoS or performance parameters such as One Way Delay (OWD), Inter Packet Delay Variation (IPDV), Packet Loss Ratio (PLR), etc.…”