<p>Because of the slow access time of disk storage, storage management is crucial to the performance of many large scale computer systems. This thesis studies performance issues in buffer cache management and disk layout management, two important components of storage management. </p><p>The buffer cache stores popular disk pages in memory to speed up the access to them. Buffer cache management algorithms used in real systems often have many parameters that require careful hand-tuning to get good performance. A self-tuning algorithm is proposed to automatically tune the page cleaning activity in the buffer cache management algorithm by monitoring the I/O activities of the buffer cache. This algorithm achieves performance comparable to the best manually tuned system.</p><p>The global data structure used by the buffer cache management algorithm is protected by a lock. Access to this lock can cause contention which can significantly reduce system throughput in multi-processor systems. Current solutions to eliminate lock contention decrease the hit ratio of the buffer cache, which causes poor performance when the system is I/O-bound. A new approach, called the multi-region cache, is proposed. This approach eliminates lock contention, maintains the hit ratio of the buffer cache, and incurs little overhead. Moreover, this approach can be applied to most buffer cache management algorithms.</p><p>Disk layout management arranges the layout of pages on disks to improve the disk I/O efficiency. The typical disk layout approach, called Overwrite, is optimized for sequential I/Os from a single file. Interleaved writes from multiple users can significantly decrease system throughput in large scale systems using Overwrite. Although the Log-structured File System (LFS) is optimized for such workloads, its garbage collection overhead can be expensive. In modern and future disks, because of the much faster improvement of disk transfer bandwidth over disk positioning time, LFS performs much better than Overwrite in most workloads, unless the disk is close to full. A new disk layout approach, called HyLog, is proposed. HyLog achieves performance comparable to the best of existing disk layout approaches in most cases.

Improvement is an essential act for restricting increasing performance pressure in organizational activities. In this paper, a framework for carrying out performance measurement process in order to strengthen the balanced score card application is developed. The proposed framework is increased to 11 procedural steps with the modification of analytic hierarchy process methods (AHP) within the balanced score card (BSC) problem solving and continuous improvement. The framework was examined with a case study of city library in Sweden. The framework enabled the service providers’ as the decision maker to assess their performance on their service rendering with respect to customer requirement, identify key measures and select the most preferred and improvement means. The motivation for this proposed framework was evident of how the step 4 can be prioritized by addressing the measure to be taken that can lead the organization in improving their performance.

Previous research has developed a number of efficient protocols for streaming popular multimedia files on-demand to potentially large numbers of concurrent clients. These protocols can achieve server bandwidth usage that grows much slower than linearly with the file request rate, and with the inverse of client start-up delay. This hesis makes the following three main contributions to the design and performance evaluation of such protocols. The first contribution is an investigation of the network bandwidth requirements for scalable on-demand streaming. The results suggest that the minimum required network bandwidth for scalable on-demand streaming typically scales as K/ln(K) as the number of client sites K increases for fixed request rate per client site, and as ln(N/(ND+1)) as the total file request rate N increases or client start-up delay D decreases, for a fixed number of sites. Multicast delivery trees configured to minimize network bandwidth usage rather than latency are found to only modestly reduce the minimum required network bandwidth. Furthermore, it is possible to achieve close to the minimum possible network and server bandwidth usage simultaneously with practical scalable delivery protocols. Second, the thesis addresses the problem of scalable on-demand streaming of a more complex type of media than is typically considered, namely variable bit rate (VBR) media. A lower bound on the minimum required server bandwidth for scalable on-demand streaming of VBR media is derived. The lower bound analysis motivates the design of a new immediate service protocol termed VBR bandwidth skimming (VBRBS) that uses constant bit rate streaming, when sufficient client storage space is available, yet fruitfully exploits the knowledge of a VBR profile. Finally, the thesis proposes non-linear media containing parallel sequences of data frames, among which clients can dynamically select at designated branch points, and investigates the design and performance issues in scalable on-demand streaming of such media. Lower bounds on the minimum required server bandwidth for various non-linear media scalable on-demand streaming approaches are derived, practical non-linear media scalable delivery protocols are developed, and, as a proof-of-concept, a simple scalable delivery protocol is implemented in a non-linear media streaming prototype system.

Previous research has developed a number of efficient protocols for streaming popular multimedia files on-demand to potentially large numbers of concurrent clients. These protocols can achieve server bandwidth usage that grows much slower than linearly with the file request rate, and with the inverse of client start-up delay. This hesis makes the following three main contributions to the design and performance evaluation of such protocols. The first contribution is an investigation of the network bandwidth requirements for scalable on-demand streaming. The results suggest that the minimum required network bandwidth for scalable on-demand streaming typically scales as K/ln(K) as the number of client sites K increases for fixed request rate per client site, and as ln(N/(ND+1)) as the total file request rate N increases or client start-up delay D decreases, for a fixed number of sites. Multicast delivery trees configured to minimize network bandwidth usage rather than latency are found to only modestly reduce the minimum required network bandwidth. Furthermore, it is possible to achieve close to the minimum possible network and server bandwidth usage simultaneously with practical scalable delivery protocols. Second, the thesis addresses the problem of scalable on-demand streaming of a more complex type of media than is typically considered, namely variable bit rate (VBR) media. A lower bound on the minimum required server bandwidth for scalable on-demand streaming of VBR media is derived. The lower bound analysis motivates the design of a new immediate service protocol termed VBR bandwidth skimming (VBRBS) that uses constant bit rate streaming, when sufficient client storage space is available, yet fruitfully exploits the knowledge of a VBR profile. Finally, the thesis proposes non-linear media containing parallel sequences of data frames, among which clients can dynamically select at designated branch points, and investigates the design and performance issues in scalable on-demand streaming of such media. Lower bounds on the minimum required server bandwidth for various non-linear media scalable on-demand streaming approaches are derived, practical non-linear media scalable delivery protocols are developed, and, as a proof-of-concept, a simple scalable delivery protocol is implemented in a non-linear media streaming prototype system.

The Internet is an essential resource to an ever-increasing number of businesses and home users. Internet access is increasing dramatically and hence, the need for efficient and effective Web server systems is on the rise. These systems are information engines that are accessed through the Internet by a rapidly growing client base. These systems are expected to provide good performance and high availability to the end user. They are also resilient to failures at both the hardware and software levels. These characteristics make them suitable for servicing the present and future information demands of the end consumer. In recent years, researchers have concentrated on taxonomies of scalable Web server system architectures, and routing and dispatching algorithms for request distribution. However, they have not focused on the classification of commercial products and prototypes, which would be of use to business professionals and software architects. Such a classification would help in selecting appropriate products from the market, based on product characteristics, and designing new products with different combinations of server architectures and dispatching algorithms. Currently, dispatching algorithms are classified as content-blind, content-aware, and Domain Name Server (DNS) scheduling. These classifications are extended, and organized under one tree structure in this thesis. With the help of this extension, this thesis develops a unified product-based taxonomy that identifies product capabilities by relating them to a classification of scalable Web server systems and to the extended taxonomy of dispatching algorithms. As part of a detailed analysis of Web server systems, generic queuing models, which consist of a dispatcher unit and a Web server unit are built. Some performance metrics, such as throughput, server performance, mean queue size, mean waiting time, mean service time and mean response time of these generic queuing models are measured for evaluation. Finally, the correctness of generic queuing models are evaluated with the help of theoretical and simulation analysis. / May 2005

<p>Because of the slow access time of disk storage, storage management is crucial to the performance of many large scale computer systems. This thesis studies performance issues in buffer cache management and disk layout management, two important components of storage management. </p><p>The buffer cache stores popular disk pages in memory to speed up the access to them. Buffer cache management algorithms used in real systems often have many parameters that require careful hand-tuning to get good performance. A self-tuning algorithm is proposed to automatically tune the page cleaning activity in the buffer cache management algorithm by monitoring the I/O activities of the buffer cache. This algorithm achieves performance comparable to the best manually tuned system.</p><p>The global data structure used by the buffer cache management algorithm is protected by a lock. Access to this lock can cause contention which can significantly reduce system throughput in multi-processor systems. Current solutions to eliminate lock contention decrease the hit ratio of the buffer cache, which causes poor performance when the system is I/O-bound. A new approach, called the multi-region cache, is proposed. This approach eliminates lock contention, maintains the hit ratio of the buffer cache, and incurs little overhead. Moreover, this approach can be applied to most buffer cache management algorithms.</p><p>Disk layout management arranges the layout of pages on disks to improve the disk I/O efficiency. The typical disk layout approach, called Overwrite, is optimized for sequential I/Os from a single file. Interleaved writes from multiple users can significantly decrease system throughput in large scale systems using Overwrite. Although the Log-structured File System (LFS) is optimized for such workloads, its garbage collection overhead can be expensive. In modern and future disks, because of the much faster improvement of disk transfer bandwidth over disk positioning time, LFS performs much better than Overwrite in most workloads, unless the disk is close to full. A new disk layout approach, called HyLog, is proposed. HyLog achieves performance comparable to the best of existing disk layout approaches in most cases.

Outsourcing is a popular name in the employ market. It is a common situation for enterprises to adopt the outsourcing system instead of hiring employees in order to achieve the goals of cost-down and elevate the competency of enterprises. Especially at the lower level technical and heavy physical- loaded industries. China Steel Corporation (CSC) has been experiencing of applying the outsourcing system on site-work over thirty years. CSC and the subcontractor are becoming a symbiosis relationship. Nevertheless, most of the documents focused on the research between the workforce flexibility and organization performance of the subcontract system, it¡¦s rare to investigate the integral job performance which are resulted from the cooperative working model in specific factory. Moreover, there are problems to be improved during the long period operation experiences under the similar symbiotic structure, such as work safety, quality, efficiency, management of manpower etc. which lead to results that are unsatisfied working performances. But there is no actual performance evaluation system to verify. This research conducts to find out the key successful factors and construct the key performance index of mechanical maintenance for job performances evaluation. It can be utilized to promote continuously for synchronizing maintenance performances of CSC and subcontractors. Eventually, enterprises¡Boutsourcing employees and subcontractors can get the best of benefits from each other. This research applies the method of Analytic Hierarchy Process (AHP) to conduct interviews from experts and questionnaires survey. It was capable to be calculated the key performance factors weight for the references of the performance evaluation standard. Furthermore, the key points of recognition of maintenance performances from supervisors were analyzed the weight differentiations of sections¡Bpositions¡Bfield experiences from the EXCEL statistic program. The final conclusions were: 1 The priority to weight Key Performance Indexes of mechanical maintenance on main hierarchy structure are management of manpower(0.346), equipment maintenance(0.27), preparation of material(0.148), working environment(0.122) , working method planning(0.113)¡C 2 The top 80 % of KPI¡¦s weight include 14 items listed as below: 2.1 promotion of reliability of equipment maintenance. 2.2 reduce the rate of resignation. 2.3 promotion of numbers of qualified for the testing of technical and operation licenses. 2.4 decreasing the waiting time of materials during the repair period. 2.5 increase periodical training hours. 2.6 elevate the level of education. 2.7 reduce accident occurring rate. 2.8 reduce the times of safety working accidents 2.9 reduce the trouble shooting time. 2.10 raise the rate of presence. 2.11 reduce the break down hours of equipments. 2.12 reduce the numbers of break down equipment. 2.13 raise amounts of equipment improvement proposals. 2.14 raise amounts of operation method improvement proposals. 3 From the individual survey analysis, it shows that there are obvious differentiation related to the management of manpower¡Bequipment maintenance¡Bworking environment. 4 Based on different sections perspectives, it shows that even though most of the sections are consistent, but having diverse recognition were found. It means that only a slice difference can be observed overall. 5 Based on different positions perspectives, it shows that different levels of authority are consistent in top two KPI indexes on main structure. the others indexes have slice differences. 6 Based on different experienced perspectives, it shows that reverse result occurred between veterans and people who are not experienced for long time of work, and the former focus on the management of manpower; the latter focus on equipment maintenance.

Abstract The request of efficiency and quality of serves of Export Processing Zone Administration from the common people or factory owners are getting increasing with the age, therefore the old layer and rigid government must also learn the popular management to increase efficiency to satisfy the public request in the fast change surrounding. Now there are many performance evaluate models, but these tools are just only used in evaluate performance .The Balanced Scorecard integrates performance evaluate and strategy action to become a best tool in this fast change age , so I take the Balanced Scorecard to analysis in my study. It¡¦s measure scope includes Financial , Customer, Internal Business Process and Learning and Growth, I hope I can use this strategic management tool to build up a standard to improve all operational performance and competition advantage. In according to ¡§Export Processing Zone Administration¡¦s performance award practice program in ninety-two year¡¨ be classified into four scopes in the Balance Scorecard. And I also use integration to get together some standards. The results as follow : 1. The current achievement evaluation standard of the Export Processing Zone Administration includes Balance Scorecard four scopes, and the whole the greater part is Internal Business Process. 2. The performance evaluation purpose is as follow: to lead staff to engage in some objective, to become a improved standard for manager, to increase satisfaction. The successful factors and measures are as follow: performance shows connect with a reward system, organization staff join this, high level managers support that, according to all kinds unit attribute be set different standards, the evaluation methods must be fair and public, a target definition is clear, quantifiable , it must be connected with performance reward and promotion. 3. Administration staff place importance on Internal Business Process aspect in selected performance index, and among which the different educational background also causes different opinions .

The research aims at benchmarking apply to local government, it proof the feasibility for public department practice benchmarking and integrate TQM spirit. In management of internal activities, the institution built benchmarking performance index, at the same time, exhibited local large-scale activities to test performance of organization act the participation and suggestion institution. The research example for Tzudying District Office, and choose Kaohsiung City Government Personnel Department as benchmarking partner. The study is to inquire about benchmarking apply on the local government office's feasibility and performance in valuation, in order to seek the complete of research outcome and practical objective. It takes¡usecond-hand data analysis¡v¡B¡uquestionnaire investigation¡vto carry on the study. In the second-hand data analysis, it consults to quote the data of Kaohsiung City Government Personnel Department. In the questionnaire investigation, by investigating the satisfactory of the masses that jointing the activity, to acquire whether the organization introduce participation and suggestion institution can enhance the activities performance. We find that it is important element about the chief of department to support benchmarking on the research, and after introducing benchmarking to the local government not only help on administration development but also it has the outstanding performance on affairs. The study proposes the following suggestions: 1. Giving more power to the chief of department is the promise of moving benchmarking. 2. Public servant's performance reward institution and benchmarking suitable to carry out for reinforce to fulfill performance management. 3. It is the only way to improve performance and sustainable development for the future government organization that emphasizing on employee's professional skill and educational training.

Currently global shipping industry has from grew stage advance to mature period. The container ship capacity of supply exceeds the demand and the economy depression impacted traditional industry fiercely more competitors join the market cause freight decline, shipping industry face to a difficult position. The transportation cost raises up accompanies by sources of energy price hike, the container ship performance improvement is getting important with each passing day. The case company depends on ¡§Port and Terminal Integrated Operating System (PATIO), and use super crane, transtainer(TT), tractor, radio data terminal (RDT) and related field partner to complete container handling process. The core competence to case company is focus on standard operating process (SOP), speedy customs formality and flexible practice. From studying academic references, case company¡¦s document and practice learn its business process reengineering rely on PATIO system as a major tool to shorten core process timing, improve vessel productivity and enhance service quality. First, the essay creates a pattern contains four phases and seven procedures for running business process reengineering (BPR), realize some topics on process improvement and management from primary data and secondary data provided by case company. Second, the study is evaluated BPR performance between beginning and finish. Third, the conclusion of key successful factors (KSF) to BPR includes: (1). Build up business vision and company values. (2). High-level management support and fully participation. (3). Equipment modification properly. (4). Management Information System Improvement. (5). Cross function member teamwork. (6). The speed of business process reengineering. (7). Employee and field partner commitment with enthusiastic participation. The conclusion of business process reengineering about case company summarize as following. 1.The pattern of business process reengineering contains four phases and seven procedures. - Planning Phase ¡P Create business vision ¡P Develop business process reengineering concept - Design Phase ¡P Map out business process reengineering contents ¡P Business process reengineering evaluation and design - Implement Phase ¡P Carry out business process reengineering plan - Measurement Phase ¡P Business process reengineering performance evaluation ¡P Continuously improvement 2.Seven key successful factors for achieving business process reengineering. - Build up business vision and company values - High-level management support and fully participation - Equipment modification properly (Bomb Cart Modification) - Management Information System Improvement - Cross function member teamwork - The speed of business process reengineering - Employee and field partner commitment with enthusiastic participation 3.To provide example for the same business to carry out business process reengineering