Side_1_360

lems is also how to optimise the performance

of a network. Solving congestion is an essen-

tial part of performance improvement. Han-

dling congestion can be divided into demand

side policies (restrictive) and supply side poli-

cies (expansive).

• Solutioncontext; elaborating how to solve
  the TE problems. This includes evaluation of
  alternatives. This requires estimating traffic
  load, characterising network state, elaborating
  solutions on TE problems and setting up a set
  of control actions. The instruments relevant
  include: i) set of policies, objectives and re-
  quirements for network performance evalua-
  tion and optimisation; ii) set of tools and
  mechanisms for measurement, characterisa-
  tion, modelling and controlling traffic and
  allocation to network resources; iii) set of con-
  straints on the operating environment, network
  protocols and TE system; iv) set of quantita-
  tive and qualitative techniques and methods

for abstracting, formulating and solving TE

problems; v) set of administrative control

parameters that may be managed by a config-

uration management system; vi) set of guide-

lines for network performance evaluation,

optimisation/improvement. Traffic estimates

can be derived from customer subscription

information, traffic projections, traffic models,

and from empirical measurements. Polices for

handling the congestion problem can be cate-

gorised according to the criteria: i) response

time scale (long – weeks to months, e.g.

capacity planning; medium – minutes to days,

e.g. setting routing parameters, adjusting

Label Switched Path (LSP) design; short –

ps to minutes, e.g. packet processing of mark-

ing, queue management); ii) reactive versus

preventive; and iii) supply side (increase

available capacity, redistribute traffic flows)

versus demand side (control the offered traf-

fic).

• Implementation and operationalcontext;
  implementing the actual solutions, involving
  planning (including a priori to determine
  actions based on triggers), organisation
  (including assigning responsibilities to differ-
  ent units and co-ordinating activities), and
  execution (including measurement and appli-
  cation of corrective and perfective actions).

These context descriptions may also be looked

upon as gradually getting more precise and

closer to the implementation.

3.2 TE Process Model

A TE process model is presented in [ID_tepri].

This is depicted in Figure 6 as an iterative proce-

dure consisting of four main steps.

The first phase includes definition of control

policies. These would typically depend on a set

of inputs, like business model, network cost

structure, operating constraints, utility model

and optimisation criterion.

The second phase involves measurementsin

order to assess the conditions in the network;

network state and traffic load.

The third phase consists of analysing the net-

work state and characterising the traffic load.

A number of potential models and analysing

techniques may be relevant, for instance also

looking at the timely and spatial distribution of

the traffic load.

In the fourth phase, performance optimisationis

done. This includes a decision process selecting

and implementing a set of actions. Actions may

work on the load demand, distribution of load

and network resource configuration and capac-

response

system

demand

system

interconnected

resources

Figure 5 Elements in the
network context

define relevant

control policies

measurement from

operational network

analyse network state and

characterise traffic load

performance optimisation

of the network

business

model network

cost structure

operating

constraints utility

model

optimisation

criterion

Figure 6 TE process model