01-06-2012, 04:22 PM
Introduction of Towards Programmable Network Measurement
Introduction of Towards Programmable Network.doc (Size: 54.5 KB / Downloads: 27)
ABSTRACT:
We present ProgME, a Programmable MEasurement architecture based on a novel concept of flowset – arbitrary set of flows defined according to application requirements and/or traffic conditions. Through a simple flowset composition language, ProgME can incorporate application requirements, adapt itself to circumvent the challenges on scalability posed by the large number of flows, and achieve a better application-perceived accuracy. ProgME can analyze and adapt to traffic statistics in real-time. Using sequential hypothesis test, ProgME can achieve fast and scalable heavy hitter identification.
Existing System:
Traffic measurements provide critical input for a wide range of network management applications, including traffic engineering, accounting, and security analysis. Existing measurement tools collect traffic statistics based on some predetermined, inflexible concept of “flows”. They do not have sufficient built-in intelligence to understand the application requirements or adapt to the traffic conditions. Consequently, they have limited scalability with respect to the number of flows and the heterogeneity of monitoring applications.
Proposed System:
We present a Programmable MEasurement architecture (ProgME) that can adapt to application requirements and traffic conditions in real time. Below figure shows the major components of ProgME. Our first proposal is to use a versatile definition of flowset – arbitrary set of flows – as the base of traffic statistics collection. In other words, ProgME keeps one counter per flowset. Compared to per-flow traffic statistics, per-flowset statistics enables one to achieve multiple resolutions within a traffic profile. Since flowsets can be defined arbitrarily, they do not necessarily map to the same number of unique flows or traffic volume. Therefore.
Modules Description:
1. Flow Set
Definition of flowset – arbitrary set of flows – as the base of traffic statistics collection. In other words, ProgME keeps one counter per flowset. Compared to per-flow traffic statistics, per-flowset statistics enables one to achieve multiple resolutions within a traffic profile. Since flowsets can be defined arbitrarily, they do not necessarily map to the same number of unique flows or traffic volume.
2. Collect and Report Statistics:
During the measurement process, FQAE performs trafficaware optimization by sorting the order of candidates in the table of matching candidates based on the number of packets observed earlier (TrafficSort). Note that this seemingly simple optimization is possible only because FQAE make flowsets fully disjoint. If flowsets have non-empty intersections, finding the optimal order is NP-complete, and one will have to resolve to heuristics, as some have attempted in the context of packet Filtering .