Print version

Network control and management architectural framework supporting military quality of service

• Download this publication

Abstract

The growing demand for multimedia/ multiparty services and their rapid design, deployment and modification using cell/packet infrustructure is apparent in the ADF. A well-chosen uniform Military Network Control and Management (M-NC&M) architecture covering both strategic and tactical domains may have an enormous impact on achieving these goals. This report proposes a general framework for such an architecture, in the medium term (5-10 years). The proposed framework follows concepts of an open and programmable network environment, and adds to them military specifics pertinent to the ADF. A particular emphasis of the framework is put on the way the military aspects of Quality of Service should be supported by the M- NC&M.

Executive Summary

The growing demand for multimedia/multiparty services and their rapid design, deployment and modification is apparent in the ADF. To fulfil this requirement a high bandwidth (i.e. hundreds of Mbit/ s) fixed core network with medium bandwidth (i.e. tens of Mbit/ s) mostly satellite links to the tactical domain are being built. Packet/cell switching technologies will be used in this interconnected environment due to their more flexible bandwidth allocation, simpler control and better adaptability to bursty traffic than circuit switching counterparts. To achieve good prediction of multimedia service behaviour, end-to-end Quality of Service (QoS) guarantees through all crossed networks must be supported. In addition, in military networks, when not enough capacity is available to send all information (e.g. due to congestion in impoverished links, partially destroyed networking infrastructure, jamming), messages carrying mission critical information should be delivered first. Unfortunately, today's packet/cell based ADF networks do not enable traffic prioritisation based on its military importance. A well-chosen uniform Military Network Control and Management (M-NC&M) architecture covering both strategic and tactical domains may have an enormous impact on: .Ability to support end-to-end QoS; .Ease of designing, deploying and modifying new services across all interconnected assets; .Efficient use, from the military viewpoint, of these assets in stressed (overloaded) situations; .Smooth evolution of legacy capabilities; .Reduction in complexity. The aim of this report is to propose a general framework for such an architecture. It focuses on packet/cell switching technologies in the strategic communication infrastructure and its interconnection to the tactical domain. The report refers to medium term (5-10 years) design goals. The report assumes that Defence will manage switches/routers involved in passing military traffic. Any commercial carriers' involvement in transmission services between these switches/routers will be through permanent or semi-permanent paths set-up using management systems of the carriers. Each carrier will guarantee QoS of their transmission services. In the first part of the report the requirements for the proposed M-NC&M architecture are specified and a review of standardised network control and management architectures, both currently used and being developed, is given. The most important findings of this part are: (1) In the military environment, QoS should be managed as a function of its military value. This approach is referred to as delivering Military QoS (M-QoS). Greater resilience of mission-critical flows during overloads can be achieved using this approach; (2) There are strong arguments to gradually "step down" in QoS of less important military flows in overloaded networks, instead of releasing these flows; (3) The M-NC&M architecture should be as transparent as possible to the used transmission infrastructure since no single winning transportation technology for multimedia/ multiparty networks is expected in the commercial world for the foreseeable future; (4) Traditional Network Control and Management systems such as Signalling System No 7 (557), Telecommunication Management Network (TMN) and the Internet cannot support rapid designing, deploying and modifying of multimedia services, prioritisation of information flows according to their military value, or enabling graceful degradation of QoS. A solution to achieve these requirements is to create an open and programmable network environment. The term open means that standard Applicanon Procedure Interfaces (APIs) are used to access network elements (e.g. routers, ATM switches) and basic network services. The term programmable refers to flexibility achieved by using high level languages to (a) represent functionality of network resources by a set of software abstractions; and (b) access (manipulate), through APIs, capabilities of these resources. Considering these findings an M-NC&M architectural framework is proposed in the second part of the report. This framework follows concepts of an open and programmable network environment developed by TINA, XRM and IEEE P1520 standard initiatives, and adds to them military specifics pertinent for ADF networks. A particular emphasis of the framework is put on the way M-QoS should be supported by the M-NC&M. A case study is also presented showing the applicability of this framework to an A TM environment. The report recommends that DSTO should: .Improve liaison with Defence Information Systems Group (DISG), especially the Network Operations Centre and the Network Technical Services at Deakin, ACT; .Find detailed mechanisms to fulfil requirements posed to the M-NC&M. Regarding studies on mechanisms improving network robustness and performance, close collaboration should be established with Drs Miro Kraetzl and Peter Shoubridge, Communications Division, DSTO, to utilise their expertise in network performance and survivability; .Specify a standardised software interface between user applications being built by DSTO and the experimental M-NC&M implemented on the EXC3ITE network; .Launch a study to analyse distribution of service management algorithms for user applications implemented in ADF networks; .Launch a study to analyse scalability and performance of M-NC&M mechanisms; .Intensify experiments on the use of available commercial and experimental software to gain practical experience in building various M-NC&M mechanisms; .Constantly monitor new developments in research on open and programmable networks; .Establish liaisons with the COMET group at Columbia University, one of the most active groups conducting advanced research on open and programmable networks; .Establish liaisons with the US Air Force Laboratory at Rome, NY, the leading US Defence research unit working on military network management. The aim is to exchange information about M-NC&M architecture(s) that are currently used and planned to be used in US Defence networks.

Back to the top