Influence of Load on Reliability of Storage Area Networks

During the coronavirus pandemic, telecommuting is widely required, making remote data access grow significantly. This requires highly reliable data storage solutions. Storage area networks (SANs) are one of such solutions. To guarantee that SANs can deliver the desired quality of service, cascading failures must be prevented, which occur when a single initial incident triggers a cascade of unexpected failures of other devices. One such incident is the data loading/overloading, causing the malfunction of one device and further cascading failures. Thus, it is crucial to address influence of data loading on the SAN reliability modeling and analysis. In this work, we make contributions by modeling the effects of data loading on the reliability of an individual switch device in SANs though the proportional-hazards model and accelerated failure-time model. Effects of loading on the reliability of the entire SAN are further investigated through dynamic fault trees and binary decision diagrams-based analysis of a mesh SAN system.

An Enhanced List Based Packet Classifier for Performance Isolation in Internet Protocol Storage Area Networks

Consolidation of storage into IP SANs (Internet protocol storage area network) has led to a combination of multiple workloads of varying demands and importance. To ensure that users get their Service level objective (SLO) a technique for isolating workloads is required. Solutions that exist include cache partitioning and throttling of workloads. However, all these techniques require workloads to be classified in order to be isolated. Previous works on performance isolation overlooked the classification process as a source of overhead in implementing performance isolation. However, it’s known that linear search based classifiers search linearly for rules that match packets in order to classify flows which results in delays among other problems especially when rules are many. This paper looks at the various limitation of list based classifiers. In addition, the paper proposes a technique that includes rule sorting, rule partitioning and building a tree rule firewall to reduce the cost of matching packets to rules during classification. Experiments were used to evaluate the proposed solution against the existing solutions and proved that the linear search based classification process could result in performance degradation if not optimized. The results of the experiments showed that the proposed solution when implemented would considerably reduce the time required for matching packets to their classes during classification as evident in the throughput and latency experienced.

Optical Local/Metropolitan and Storage-Area Networks

The first generation of local/metropolitan-area networks (LANs/MANs) used copper-based media, spread out typically across a building or a campus under one autonomous administration. With the arrival of optical-fiber transmission, considerable developments took place to enhance the speed and size of these copper-based LANs and MANs, by using optical fibers over various possible topologies, such as bus, ring, and star. We begin with the earlier experiments carried out on optical LANs/MANs in various research groups, and then present some of the standardized optical LANs/MANs, such as distributed queue dual bus (DQDB), fiber-distributed digital interface (FDDI), and different versions of high-speed Ethernet (1/10/40/100 Gbps). Finally, we describe some of the storage-area networks (SANs) augmented by optical fiber transmission, where small locally networked clusters of servers and storage devices are accessed by workstations through LANs such as, HIPPI, ESCON, and fiber channel. (139 words)