Internet Traffic Measurement and QoS Monitoring

* QoS in Future Internet depends strongly on the ability of IP network operators to cope with customers demands in terms of bandwidth (BW) and parameters like jitter, delay, bit error rate (BER), that can be specified in the Service Level Agreement (SLA).
* Several projects have been developed to deal with IP traffic measurement: MOME, LOBSTER, ETOMIC,DIMES, BART, RIPE, .
* They use different techniques and provide data in different formats. MOMENT is a project designed to provide data from all these infrastuctures through a single semantic interface.As illustrate by the figure MOMENT is a middleware that enables users interfacing to several traffic measurement tools.
File:Momentintro.jpg
* MOMENT mediator will provide interoperability with the existing infrastructures and will be open for new ones to ensure enhanced functionality of the platform: The mediator architecture and the unified interface development are generic and can be extended to future tools provided some standard interfacing be agreed. It also takes into account the importance of preserving confidential data of Internet users. MOMENT is currently being developed by a set of partners from Europe in cooperation with CAIDA group from USA. Such a effort is required to perform Internet Traffic Measurement and Monitoring since:
** Users have problems to access IP traffic infrastructures and understand the way they work as well as the format of data that can be retrieved.
** Single measurements issued from different tools are difficult to be combined.
* MOMENT middleware tackles this task by means of a semantic approach.
* To do this, it's needed a measurement-specific ontology, allowing semantic representation and retrieval of measurement and monitoring information, as well as providing the flexibility of a service-oriented architecture for future Internet applications.
Architecture
* The mediator to heterogeneous measurement infrastructures and data repositories is based on a Service Oriented Architecture (SOA) approach with the help of Web-services. The ontology-based architecture allows the system to overcome the vast differences between data bases and tools.
* The platform is non-intrusive as ISPs concern: Monitoring probes are placed outside the commercial network domains, data obtained by passive monitoring are anonymised and middleware solutions has been deployed inside the system to properly control and manage access to data subject to privacy issues. But research groups can also use the mediator and flexibly design both, passive and active measurements, so that they perform their own experiments provided their user profile meets the expertise level required.
* MOMENT mediator allows the integration of all different kinds of data sources from very straightforward up to very sophisticated.
**The Unified Interface (UI) presents the facilities to record the usage of the mediator, keep a repository of its functionality and help beginners to use it as well as enable fora about its results. Besides provides access to the Mediator for both:
*** Humans through a Graphical User Interface (GUI).
****To query or browse the mediadata registry.
****To execute measurement data queries.
****To define and execute workflows.
****To set a data analysis method.
****To define a data transformation.
***Machines via web service (WS).
***To invoke the individual mediator components to execute a workflow.
** The Semantic Metadata Registry (SMR) stores information about the method to access measurement data in a mediated infraestrcuture. This information is passed on to the invocation service.
***All data in the metadata registry are based on a common ontology and therefore semantically interpretable
** The Semantic Query Integration Services (SQIS) executes queries that access infrastructures directly, allowing a fine grained search for measurement data stored in query-enabled sources.These queries are requested by an application on top of the unified interface using semantic technologies such as SPARQL and the answer is integrated from existing sources, using a common vocabulary taken from the network data ontology.
** The Analysis Service (ANAS) is responsible for further analysis of measurement results. This service can process data from one ore more infrastructure(s). Information about available analysis services is stored in the metadata registry.
** The Anonymization Service (ANOS) provides methods to anonymize measurement data in order to preserve privacy.
** The Workflow Service (WFS) helps to automate recurring processes. In a workflow a user can define a sequence of MOMENT operations
** The Invocation Service (INVS) is used when a user wants to retrieve specific measurement data from an infrastructure.
** The Invocation Service (INVS) is used when a user wants to retrieve specific measurement data from an infrastructure.
** The Infrastructure Interface (II) infrastructure interface is mainly responsible for receiving semantically enhanced metadata information from the infrastructures. It provides a web service (WS) interface to push their metadata into the system in order to make their measurement data findable and accessible. Infrastructures can publish their metadata in RDF triples which are defined in the MOMENT ontology.
Ontology standardization
* Internet traffic monitoring by means of Universal Interfaces will not be possible unless a general agreement for semantic tools the be reached. This is the goal of the MOI (Measurement Ontology for Internet) ISG (Industrial Specification Group) set in the ETSI. Other fora are being asked to cooperate in other two reach standard tools for feature Internet Traffic Monitoring as pointed out in the MANA position paper of the Future Internet Assembly (FIA).

Main Menu

See Also

Main Menu