Search

Providing adequate computing lab support for Management Information Systems (MIS) and Computer Science (CS) programs is a perennial challenge for most academic institutions in the US and abroad. Factors, such as lack of physical space, budgetary constraints, conflicting needs of different courses, and rapid obsolescence of computing technology, precipitate this challenge. While introductory computing courses can be adequately supported through general labs, advanced computing courses typically require dedicated labs. Normally, however, even a dedicated lab must simultaneously support multiple courses with diverse support requirements. Consequently, in designing dedicated computing labs, versatility and flexibility of the lab must be the prime consideration. Drawing on the insights gained from developing a dedicated lab, the authors present a model for developing a versatile, flexible lab to support computer network courses. The proposed model entails a modular design approach which offers ease of management, configuration and upgrading of the lab by decoupling the lab functions into layers. The layered design divides the lab into four management layers: (1) wiring layer, (2) networking layer, (3) server systems layer, and (4) support procedures layer. Each layer offers a versatile and configurable design and serves as a foundation for the next higher layer. Wiring layer minimizes the wire configuration efforts and enhances wiring flexibility by utilizing patch panels at the computer desks and the racks housing networking devices, and sufficient underground wiring between desks and racks. The underground wiring allows for network, telephone and serial cable connections as needed. Network layer builds LANs, subnets and internets with minimal configuration effort through a careful arrangement of network hubs, switches, routers, and firewalls. The network layer distinguishes between permanent networks needed in the lab, called pre-configured networks, and the networks that are assembled and disassembled by the students, called configurable networks. The network address allocation schemes for different types of network experiments are also discussed. Server systems layer covers two categories of servers: pre-configured servers and configurable servers. Pre-configured servers, such as DNS and file servers, are production servers that support network experiments. Configurable servers are hardware platforms that can be configured as needed according to experimental needs. The last layer, procedural layer, provides policies, guidelines and operational procedures that are needed for proper utilization of the lab space and the equipment. The paper also illustrates the application of the proposed model through the design and configuration of the computing lab for network courses underlying this case study. [ABSTRACT FROM PUBLISHER]

Budgetary constraints are almost a constant challenge confronted by academia. In the case of small colleges and universities, these constraints become even more pronounced when these institutions enhance their programs to meet the needs of their constituents. While in some cases program enhancement might simply be accomplished by hiring new faculty members; in others, it might require additional resources. This is especially true for computer science or management information systems programs; the changes or enhancements in these programs also mean additional computing lab support. However, budgetary constraints, lack of physical space to house the labs, conflicting needs of different courses, and rapid obsolescence of technology only exacerbate this dilemma. Needless to say, overcoming these challenges requires some pragmatic thinking and approaches. Based on their own experiences in dealing with these same challenges, the authors suggest the following guidelines for configuring computing labs that are cost-effective, flexible and versatile: * Categorize computinbudgg courses, based on their computing support requirement, into groups such as general computing, advanced computing, and special projects courses. * As far as possible, support general computing courses through open/general labs and, thereby, minimize the demand for dedicated/special computing labs. * The advanced computing courses will likely require dedicated labs, due to unique software and/or hardware requirements. For most institutions, creating a dedicated lab for every course is infeasible. Consequently, design a lab that can support multiple courses by use of removable hard drives and flexible lab configurations. * If the physical space is not available to house a dedicated computer lab, a classroom may be networked into an isolated laptop-computing lab. To further enhance the flexibility of this lab, it may be connected to the institution's computing networks through wireless network configuration. * For special projects, develop a dedicated computing lab with highly flexible configuration and infrastructure to accommodate the unique computing requirements of different projects. Furthermore, this lab should mimic the network and organizational configuration of a typical organization to impart real-life flavor to special projects. * With proper supervision, policies and procedures, student personnel may be employed to maintain and manage specialized labs. This will help contain personnel costs and at the same time furnish valuable experience to the student personnel. These guidelines should help small U.S., as well as international, institutions that are planning or are in the process of configuring computing labs to support their computer science and management information systems programs. [ABSTRACT FROM AUTHOR]

Today is an era where multiprocessor technology plays a major role in designs of modern computer architecture. While multiprocessor systems offer extra computing power, it also opens a new range of opportunities to improve fault-robustness. This paper focuses on a problem of achieving fault-tolerance using replications in real-time, multiprocessor systems. In the problem, multiple replicas, or copies, of a computing task are executed on distinct processors to resist potential processor failures and computing faults. Two greedy, approximation heuristics, named Worst Fit Increasing K-Replication and First Fit Increasing K-Replication, are studied to maximise the number of real-time tasks assigned on a system with identical processors, respecting to the tasks’ replicating and timely requirements. Worst case performance is analysed by using an approximation ratio between the algorithms and an optimal solution. We mathematically prove that the ratios of using both algorithms are infinitely close to 2. Simulations are performed on a large set of testing cases which can be used to bring to light the average performance of using the algorithms in practice. The results show that both heuristic algorithms provide simple but fast and effective solutions to solve the problem. Assigning real-time tasks to a multiprocessor system with replications. [ABSTRACT FROM PUBLISHER]