T. Brothers,∗ N. Mandagere,∗ S. Muknahallipatna,∗ J.C. Hamann,∗ and H. Johnson∗∗
Performance evaluation, mass storage, IP storage, topology, distance, simulation
The ever-increasing demand for fast and reliable data storage has caused dramatic changes in the storage model. Direct attached storage (DAS) architecture, also known as server-centric architec- ture, is being replaced by the storage area network (SAN), leading to storage-centric architecture. In SAN storage devices exist com- pletely independent of any computer (servers in particular). Multi- ple servers can access the same storage device without involvement of other servers. Recently, companies have been able to implement storage integration/sharing, remote backup/restoral, near-real-time disk mirroring/replication, and the like, using a new SAN architec- ture known as distributed storage area network (DSAN). The DSAN is multiple SANs interconnected using existing Internet Protocol (IP) networks to distribute storage over wide geographical areas. Even though DSANs provide additional capabilities to a SAN, the end-to-end latency (ETEL) due to large physical cable lengths (IP networks) imposes limitations on enterprise class application perfor- mance. DSANs can also be implemented using IP storage solutions that provide direct links between servers and storages, eliminating translation of fibre channel protocol frames to IP packets. This paper examines the effect of ETEL in DSAN on the performance of an enterprise class application with real-time application replication like Microsoft Exchange Server 2003. This investigation in done in two parts: disk subsystem (DS) characterization using Jetstress benchmarking software and the performance of Microsoft Exchange using LoadSim 2003 benchmarking software. Preliminary results were presented in [1].
Important Links:
Go Back
