Integrated QOS management technique for internet protocol storage area networks
Abstract
The increasing number of Information technology Users around the world has
led to tremendous increase in the amount of data that requires storage. In
response to this challenge, new storage area network architectures based on
Ethernet (IP) have evolved. With the coexistence of storage traffic with other
types of traffic in the same IP network, it is important to offer storage traffic
QOS guarantees to prevent performance degradation for storage users.
Regrettably, the storage device itself does not provide any capability of
guaranteeing storage QOS. QOS is a vital issue in environment of mixed works
like IP SANS. The main aim of the study was to analyses the QOS techniques
used in IP networks, design, develop and validate an Integrated QOS
management technique for IP SANs. The study first analyzed the various
techniques for achieving QOS in IP Networks. By decomposing QOS problem
into an integration of four techniques of performance isolation, bandwidth
management and burst handling the study designed and developed IQMIS, an
integrated quality of service management technique for IP SANS. The study
adopted experimental research design. Simulations were used as the source of
data where Park dale tool was used for simulating reads and writes to the targets.
The study generated quantitative results which were analyzed using descriptive
statistics and results presented in tables and charts. Empirical results show that
IQMIS enables users to fairly share the aggregate system throughput even in
environment of contention of resources with a small implementation cost of 6%.
In the implementation of bandwidth management and burst handling, IQMIS
was found to be work conserving and quickly adopts to network changes with a
convergence time of 10 seconds. Further the results show that IQMIS can
provide strong performance isolation, superior latency, throughput and jitter
compared to best effort. Ultimately IQMIS can be used to provide end to end
QOS management in IPSANS and at the same time provide building blocks for
providing QOS in IPSANS.to tremendous increase in the amount of data that
requires storage.