Most IS organizations currently lack the basic information that they need to get the most value from their distributedcomputing assets. Many do not even know how many workstations and servers are in their enterprise....
详细信息
Most IS organizations currently lack the basic information that they need to get the most value from their distributedcomputing assets. Many do not even know how many workstations and servers are in their enterprise. As more client/server applications are deployed, distributed computer management becomes much more complex than simple network management. This paper will ″clear the fog″ by describing how a distributed asset repository can help effectively manage client/server systems. Maintaining up-to-date information about the location, configuration, and status of all distributed assets is the first step. IS management, finance, client/server application developers, the help desk, network planners, and operations need this information to effectively manage their distributedcomputing enterprise. By collecting up-to-date information and tracking assets as they change, users not only know the present state of the asset and the history, but can also use forecasting to make better business decisions.
This paper reviews the purpose of Software Performance Engineering (SPE) and the steps for applying the SPE methods throughout the life cycle of new systems. It describes the data required for SPE studies and illustra...
详细信息
This paper reviews the purpose of Software Performance Engineering (SPE) and the steps for applying the SPE methods throughout the life cycle of new systems. It describes the data required for SPE studies and illustrates a performance walkthrough for gathering the data. It reviews the software and system execution models for assessing the performance of alternatives and for capacity planning to support the new applications.
Future distributed systems are expected to support mobile computing through wireless extensions. In such environments, the spatial location of system resources, such as servers, can change dynamically as a result of h...
详细信息
ISBN:
(纸本)9780897918725
Future distributed systems are expected to support mobile computing through wireless extensions. In such environments, the spatial location of system resources, such as servers, can change dynamically as a result of host mobility. Consequently, a search or discovery cost is incurred whenever a client requests use of a mobile resource whose location is unknown. In addition, because of the acute dependence of radio link performance on the characteristics of indoor operating environments, the Quality of Service experienced by a client that utilizes a mobile resource can depend on the resource's specific location. The work reported in this paper proposes and evaluates a statistical framework for resource discovery in distributedcomputing environments with mobile hosts. Specifically, taking the peculiar attributes of wireless segments of the distributed system into account, we investigate the interplay between the cost of search for potentially mobile resources and the utility or benefit accrued to clients that require such resources. This interplay has important ramifications for resource allocation in distributed mobile computing environments.
The rapid introduction of new computing technologies into the enterprise frequently requires the evaluation of architectural alternatives with only minimal performance and capacity data available. Logical and physical...
详细信息
distributed systems consist of many components such as servers, clients, networks, CPU's, disks, processes and database instances. These components interact in complex ways. The performance management of a distrib...
详细信息
distributed systems consist of many components such as servers, clients, networks, CPU's, disks, processes and database instances. These components interact in complex ways. The performance management of a distributed system is complicated by the number of components, their diversity, their inter-relationships, configuration changes, and rapid changes in technology. The problem is further compounded by the ever decreasing number of technical professionals available for performance monitoring. We describe an object-oriented approach to managing the performance of such systems. We further explain how the object-oriented approach, through the use of inheritance and containment, allows one to manage these systems via well defined policies that can be easily enhanced, distributed and applied.
Object-orientation increases portability, flexibilty and reuse of software components;distribution improves performance, scalability and collaboration. We propose a distributed object computing (DOC) framework for the...
详细信息
This paper surveys the critical issues facing distributed network management researchers. The explosive growth in the use of the World Wide Web adds urgency to the need for designers of enterprise management platforms...
详细信息
This paper surveys the critical issues facing distributed network management researchers. The explosive growth in the use of the World Wide Web adds urgency to the need for designers of enterprise management platforms to add distributed management functionality to available network management products. After enumerating evaluation criteria for judging the merits of current network management systems, the report discusses the key issues which present challenges to the design of future network management tools. Realizing the significance of security to widespread adoption of network management products, this paper focuses on the proposed security models for SNMPv2 currently being considered by SNMP researchers. By including a discussion of the differences between the SNMPv2* and USEC models, the intent is to demonstrate the importance of the implementation of new SNMP security features to the future role of distributed network management platforms.
The rapid introduction of new computing technologies into the enterprise frequently requires the evaluation of architectural alternatives with only minimal performance and capacity data available. Logical and physical...
详细信息
The rapid introduction of new computing technologies into the enterprise frequently requires the evaluation of architectural alternatives with only minimal performance and capacity data available. Logical and physical architectures for technologies such as distributed management, corporate intranets, workgroup servers, and DCE/Encina software infrastructure must be quickly designed and evaluated for soundness. This paper discusses the performance engineering methodology used to rapidly evaluate the physical architecture that was proposed for the deployment of DCE and Encina infrastructure at U S WEST. One of the critical concerns about this architecture was its ability to support the required service levels, and no published guidelines were available. The engineering model that was developed to support the methodology is general enough in its approach that it is currently being modified for performance engineering work being done for Windows NT and for FORTE-based applications.
This paper outlines the operational needs of software products that manage batch workloads and associated resource utilization in systems of distributed shared computers. An approach to developing product requirements...
详细信息
This paper outlines the operational needs of software products that manage batch workloads and associated resource utilization in systems of distributed shared computers. An approach to developing product requirements based on representative enterprise models is presented. The approach identifies needs for next generation systems management tools in areas such as dynamic resource management, policy administration, and strategic utilization management. These concepts reflect the collaboration of three technology providers and their interactions with high profile enterprises in need of more effective tools for managing computer resources. Product developers and systems managers should find the material to be helpful in planning future products and product acquisitions. Equally important, the paper illustrates the value of integrating organizational models into the strategic requirements planning process.
Today's complex distributed client/server systems present a new challenge to the performance analyst. At the core of all performance analysis methodologies is consistent and reliable statistical data, gathered fro...
详细信息
Today's complex distributed client/server systems present a new challenge to the performance analyst. At the core of all performance analysis methodologies is consistent and reliable statistical data, gathered from each of the critical components of the system. In the mainframe world, we rely heavily on SMF as a data source. There is no such convenient and ready data source as SMF for the distributed environment. SNMP, and it's RMON extension, are industry wide standards that can provide the necessary data;in fact, the data is already being collected. However, unlike SMF, using SNMP to create a performance database presents a unique set of challenges. Overcoming these challenges requires a very different approach to gathering the data. This paper provides an overview of what SNMP is, how it works and how to create a performance database with SNMP.
暂无评论