checks the threshold at first, and compares the
metrics value secondly. The threshold is the value
which can be alerted to the operator by sending the
‘warning’. If the operator receives the warning
message from the system, he/she reviews the details,
and can take an action to prevent the violation of
SLA.
The MM has the following monitoring
functions : service opening, trouble, and
performance monitoring. At the system initiation
stage, the MM creates three threads in order to
monitor the categorized metrics.
At the defined thread invoke time, the MM
periodically creates threads. Threads retrieve the
monitored data, threshold and metric. Firstly, the
MM thread compares the data with the threshold. If
the current value is greater than the threshold, the
‘warning’ message is sent to the AM via XML
format. The MM thread will detect the violation of
SLA by comparing the issued time with metric value.
If the violation event occurs, the ‘violation’ message
will be sent, and the violation details is recorded in
the database through the DM. Finally, the threads
are disposed after execution.
Policy-based monitoring can be accomplished by
configuring the various preferences. The interval of
monitoring can be changed by using the UM. If the
operator changes the monitoring interval, the UM
sends the message to the MM. The MM receives the
event, and changes the thread invoke time.
Furthermore, the operator can have the metrics
monitored or not. Using the UM, the operator can
configure whether the metrics are monitored or not.
If the metric is set not to be monitored, the MM will
not execute the monitoring function. But if the
history of warning and violation is recorded in the
database, and can be retrieved via the UM.
5 CONCLUSIONS
We propose a form of architecture for an SLA
Management System using web service. We first
explore the SLA-related works in various network
environments. As describing the characteristics of
WLAN, we issued the metrics related WLAN. A
framework for SLM over WLAN are introduced
with detailed description of its components. Our
system has the capability to manage the SLA from
the service opening to the service termination. By
using the client-side agent which collects the
network performance information, we can easily
manage the SLA, and control the QoS.
According to the implications of the research,
future work has been conducted to interwork
Operation Supporting Systems (OSSs) such as the
refund system and NMS.
REFERENCES
F. De Turck et al., 2001. Design and Implementation of a
Generic Connection Management and Service Level
Agreement Monitoring Platform Supporting the
Virtual Private Network Service, Integrated Network
Management Proceedings.
Verma, D.C., 2004. Service Level Agreements on IP
Networks, Internet Computing, IEEE.
Fawaz, W., Daheb, B., Audouin, O., Du-Pond, M., Pujolle,
G., 2004. Service level agreement and provisioning in
optical networks, Communications Magazine, IEEE
K. Trzec, B. Mikac, 2004. On Agent-based Service Level
Agreement Management in Optical Internet,
MELECON, IEEE.
R. Chakravorty, I. Pratt, J. Crowcroft, 2003. A framework
for dynamic SLA-based QoS control for UMTS,
Wireless Communications, IEEE.
M. Buco, Rong Chang, L. Luan, C. Ward, J. Wolf, P. Yu,
2003. Managing eBusiness on demand SLA contracts
in business terms using the cross-SLA execution
manager SAM, ISADS.
M. D’Arienzo, M. Esposito, S.P. Romano, G. Ventre,
2003. Automatic SLA Management in SLA-aware
architecture, Telecommunications, 2003 (ICT 2003).
G. Cortese, R. Fiutem, P. Cremonese, S. D'antonio, M.
Esposito, S.P. Romano, A. Diaconescu, 2003.
CADENUS: creation and deployment of end-user
services in premium IP networks, Communications
Magazine, IEEE.
Acharya, A., Bisdikian, C., Misra, A., Young-Bae Ko.,
2004. airConn: A framework for tiered services in
public wireless LAN hot spots, Communications
Magazine, IEEE
ICETE 2005 - GLOBAL COMMUNICATION INFORMATION SYSTEMS AND SERVICES
176