Abstract
In order for advanced applications in modern computer networks to function satisfactorily, there is often the need for a guaranteed network performance and guaranteed values for several network parameters. When the provisioning and usage of network services is agreed, relevant specifications for the level of the services are also defined.
Key Terms in this Chapter
Capacity: In the context of this article, capacity is the bandwidth available for a specific service. The exact definition of the capacity (for example, at which network layer it should be measured) is a subject of the SLA for the service.
Quality of Service (QoS): The ability to provide specific guarantees to traffic flows regarding the network characteristics such as packet loss, delay, and jitter experienced by the flows.
Availability: In the context of this article, availability is the percentage of time that a specific service is operating satisfactorily. The exact definition of the availability (for example, under which circumstances the service is no longer considered available) is a subject of the SLA for the service.
SLA: Service level agreement—a formal agreement between a Service Provider and a Service Client to provide a service with well-defined service levels, accompanied by possible penalty clauses if the SLA is not met.
Service Response Time: Service response time is the time from the moment that a user makes a request to the server, until the result of the request comes back.
Compensations: The possible actions (such as payments or forfeiture of a fee) against a service provider that fails to meet the level of service defined in the SLA.
Metrics: The characteristics that are properly welldefined so that they can be measured and used to evaluate the service and its compliance to the SLA.