SDN-based Traffic Matrix Estimation in Data Center Networks Through Large Size Flow Identification


In data center networks, software-defined NetWorking (SDN), which has a control plane that is distinct from the data plane, enables the creation of new opportunities for traffic measurement. However, the TCAM (Ternary Content Addressable Memory) resources that are available for traffic measurement in switches that support SDN are severely constrained. Therefore, it is necessary to utilize traffic matrix (TM) estimation in order to derive a hybrid network monitoring scheme by combining the partial direct measurement provided by SDN with some inference techniques. This can be done by combining the two sets of data. Directly monitoring each flow and discovering that large size flows consume a massive amount of channel bandwidth resource between control plane and data plane may reveal that large size flows play an important role in improving the accuracy of TM estimation; however, this discovery can only be made by monitoring each flow individually. Instead of monitoring each flow individually, as a result, in this paper we focus on identifying large size flows based on multiple historical TMs. First, we perform an analysis on multiple historical TMs and notice that an origin-to-destination (OD) pair whose flow size was chosen as a large size flow during the previous time slot is most likely to be chosen for per-flow monitoring during the following time slot. Because of this, these OD pairs are identified by a gradient boosting machine and are directly regarded as sampled OD pairs in order to reduce the amount of resources that are used. Then, we present a greedy heuristic algorithm as a solution to the problem of selecting SDN-enabled switches. This will allow us to make the most efficient use of the TCAM resources and ensure that the majority of the sampled OD pairs are accounted for in the flow table. We also present a source node prefix tree based bit merging aggregation (SPTBMA) scheme to design feasible forwarding rules to be inserted in TCAM of SDN-enabled switches and reserve more TCAM space for sampled OD pairs. This scheme was developed in order to design feasible forwarding rules to be inserted in TCAM of SDN-enabled switches. Finally, the experimental results that were based on a real traffic dataset demonstrated that our proposed scheme outperforms the existing algorithms in terms of improving the accuracy of TM estimation and overcoming the limitation of TCAM resources.

Did you like this research project?

To get this research project Guidelines, Training and Code... Click Here

PROJECT TITLE : Systematic Analysis of Fine-Grained Mobility Prediction with On-Device Contextual Data ABSTRACT: The concept of predicting the mobility of users is widely discussed within the research community. Numerous studies
PROJECT TITLE : Objective-Variable Tour Planning for Mobile Data Collection in Partitioned Sensor Networks ABSTRACT: Wireless sensor networks can achieve greater energy efficiency and more even load distribution through the collection
PROJECT TITLE : Location-Flexible Mobile Data Service in Overseas Market ABSTRACT: Mobile network operators, also known as MNOs, are the companies that are responsible for providing wireless data services. These services are based
PROJECT TITLE : Parallel Fractional Hot-Deck Imputation and Variance Estimation for Big Incomplete Data Curing ABSTRACT: The fractional hot-deck imputation, also known as FHDI, is a method for handling multivariate missing data
PROJECT TITLE : Representation Learning from Limited Educational Data with Crowdsourced Labels ABSTRACT: It has been demonstrated that representation learning plays a significant part in the unprecedented success of machine learning

Ready to Complete Your Academic MTech Project Work In Affordable Price ?

Project Enquiry