Civil service reform trends in public personnel management: Experiences from the Commonwealth of the Bahamas

This study has explored the potential for implementing a merit-based public personnel system in The Bahamas, a former British colony in The Commonwealth Caribbean. Specifically, the study evaluated the use of merit-based public personnel management practices in areas of recruitment, selection, promotion, training and employee development and performance evaluation. Driving forces and barriers which impact merit system successes and failures as well as strategies for institutionalizing merit system practices are identified. Finally the study attempted to apply the developmental model created by Klingner (1996) to describe the stage of public personnel management in The Bahamas. The data for the study was collected through in-depth interviews with expert observers. ^

Economic assessment of best management practices in the Mara River Basin : toward implementing payment for watershed services

The Mara River in East Africa is currently experiencing poor water quality and increased fluctuations in seasonal flow. This study investigated technically effective and economically viable Best Management Practices for adoption in the Mara River Basin of Kenya that can stop further water resources degradation. A survey of 155 farmers was conducted in the upper catchment of the Kenyan side of the river basin. Farmers provided their assessment of BMPs that would best suit their farm in terms of water quality improvement, economic feasibility, and technicalsuitability. Cost data on different practices from farmers and published literature was collected. The results indicated that erosion control structures and runoff management practices were most suitable for adoption. The study estimated the total area that would be improved to restore water quality and reduce further water resources degradation. Farmers were found to incur losses from adopting new practices and would therefore require monetary support.

Recruiting and developing technicians for hotel food service operations

The thesis which follows is a study of recruiting and developing skilled workers for Hotel Food Service Operations in the Miami area. The aim of the study is to bring to the attention of personnel management the role of recruiting and training in providing the skilled people needed for their operation in the short and long run as well. The study was done as a case study of the medium and large size hotels which have a minimum of 250 units each in the Miami area. However, the study has been generalized where it is possible, and when data permitted. The primary data was collected by the use of the questionnaire survey method composed of key questions about recruiting, training and sources of skilled people, turnover reasons, etc. Eight tables have been constructed, analyzed and interpreted. A personal opinion was mentioned in the interpretation of each table's data. It was found that personnel management should provide a better recruiting and developing procedures in order to attract more qualified people, particularly among the youngsters who are potential skilled workers for the future. It was concluded that the quality of work life, the benefits, and the opportunities for advancement in the food and beverage operations play a significant role in an employee's decision to stay with a particular job, and to acquire the necessary skills.

Integrated assessment of water conservation practices for sustainable management strategies

Miami-Dade County implemented a series of water conservation programs, which included rebate/exchange incentives to encourage the use of high efficiency aerators (AR), showerheads (SH), toilets (HET) and clothes washers (HEW), to respond to the environmental sustainability issue in urban areas. This study first used panel data analysis of water consumption to evaluate the performance and actual water savings of individual programs. Integrated water demand model has also been developed for incorporating property’s physical characteristics into the water consumption profiles. Life cycle assessment (with emphasis on end-use stage in water system) of water intense appliances was conducted to determine the environmental impacts brought by each practice. Approximately 6 to 10 % of water has been saved in the first and second year of implementation of high efficiency appliances, and with continuing savings in the third and fourth years. Water savings (gallons per household per day) for water efficiency appliances were observed at 28 (11.1%) for SH, 34.7 (13.3%) for HET, and 39.7 (14.5%) for HEW. Furthermore, the estimated contributions of high efficiency appliances for reducing water demand in the integrated water demand model were between 5 and 19% (highest in the AR program). Results indicated that adoption of more than one type of water efficiency appliance could significantly reduce residential water demand. For the sustainable water management strategies, the appropriate water conservation rate was projected to be 1 to 2 million gallons per day (MGD) through 2030. With 2 MGD of water savings, the estimated per capita water use (GPCD) could be reduced from approximately 140 to 122 GPCD. Additional efforts are needed to reduce the water demand to US EPA’s “Water Sense” conservation levels of 70 GPCD by 2030. Life cycle assessment results showed that environmental impacts (water and energy demands and greenhouse gas emissions) from end-use and demand phases are most significant within the water system, particularly due to water heating (73% for clothes washer and 93% for showerhead). Estimations of optimal lifespan for appliances (8 to 21 years) implied that earlier replacement with efficiency models is encouraged in order to minimize the environmental impacts brought by current practice.

Improving resource management in virtualized data centers using application performance models

The rapid growth of virtualized data centers and cloud hosting services is making the management of physical resources such as CPU, memory, and I/O bandwidth in data center servers increasingly important. Server management now involves dealing with multiple dissimilar applications with varying Service-Level-Agreements (SLAs) and multiple resource dimensions. The multiplicity and diversity of resources and applications are rendering administrative tasks more complex and challenging. This thesis aimed to develop a framework and techniques that would help substantially reduce data center management complexity.^ We specifically addressed two crucial data center operations. First, we precisely estimated capacity requirements of client virtual machines (VMs) while renting server space in cloud environment. Second, we proposed a systematic process to efficiently allocate physical resources to hosted VMs in a data center. To realize these dual objectives, accurately capturing the effects of resource allocations on application performance is vital. The benefits of accurate application performance modeling are multifold. Cloud users can size their VMs appropriately and pay only for the resources that they need; service providers can also offer a new charging model based on the VMs performance instead of their configured sizes. As a result, clients will pay exactly for the performance they are actually experiencing; on the other hand, administrators will be able to maximize their total revenue by utilizing application performance models and SLAs. ^ This thesis made the following contributions. First, we identified resource control parameters crucial for distributing physical resources and characterizing contention for virtualized applications in a shared hosting environment. Second, we explored several modeling techniques and confirmed the suitability of two machine learning tools, Artificial Neural Network and Support Vector Machine, to accurately model the performance of virtualized applications. Moreover, we suggested and evaluated modeling optimizations necessary to improve prediction accuracy when using these modeling tools. Third, we presented an approach to optimal VM sizing by employing the performance models we created. Finally, we proposed a revenue-driven resource allocation algorithm which maximizes the SLA-generated revenue for a data center.^

Loaxahatchee Impoundment Landscape Assessment (LILA): Tree Island Experiments and Management; May 1, 2005 to September 4, 2009: Final Report

Hydrologic modifications have negatively impacted the Florida Everglades in numerous significant ways. The compartmentalization of the once continuously flowing system into the Water Conservation Areas (WCAs) caused disruption of the slow natural flow of water south from Lake Okeechobee through the Everglades to Florida Bay. The ponding of water in the WCAs, the linking of water flow to controlled water levels, and the management of water levels for anthropogenic vs. ecological well-being has caused a reduction in the spatial heterogeneity of the Everglades leading to greater uniformity in topography and vegetation. These effects are noticeable as the degradation in structure of the Everglades Ridge and Slough environment and associated Tree Islands. In aquatic systems water flow is of fundamental importance in shaping the structure and function of the ecosystem. The organized patterns of parallel orientation of ridges, sloughs, and tear-drop shaped tree islands along historic flow paths attest to the importance of water movement in structuring this system. Our main objective was to operate and manage the LILA facility to provide a broad potential as a research platform for an integrated group of multidisciplinary, multi-agency scientists collaborating on multifunctional studies aimed primarily at determining the effects of CERP water management scenarios on the ecology of tree islands and ridge and slough habitats. We support Everglades water management, CERP, and the Long-Term Plan by defining hydrologic regimes that sustain healthy tree islands and ridge and slough ecosystems. Information gained through this project will help to reduce the uncertainty of predicting the tree island and ridge and slough ecosystem response to changes in hydrologic conditions. Additionally, we have developed the LILA site as a visual example of Everglades restoration programs in action.

Improving Resource Management in Virtualized Data Centers using Application Performance Models

The rapid growth of virtualized data centers and cloud hosting services is making the management of physical resources such as CPU, memory, and I/O bandwidth in data center servers increasingly important. Server management now involves dealing with multiple dissimilar applications with varying Service-Level-Agreements (SLAs) and multiple resource dimensions. The multiplicity and diversity of resources and applications are rendering administrative tasks more complex and challenging. This thesis aimed to develop a framework and techniques that would help substantially reduce data center management complexity. We specifically addressed two crucial data center operations. First, we precisely estimated capacity requirements of client virtual machines (VMs) while renting server space in cloud environment. Second, we proposed a systematic process to efficiently allocate physical resources to hosted VMs in a data center. To realize these dual objectives, accurately capturing the effects of resource allocations on application performance is vital. The benefits of accurate application performance modeling are multifold. Cloud users can size their VMs appropriately and pay only for the resources that they need; service providers can also offer a new charging model based on the VMs performance instead of their configured sizes. As a result, clients will pay exactly for the performance they are actually experiencing; on the other hand, administrators will be able to maximize their total revenue by utilizing application performance models and SLAs. This thesis made the following contributions. First, we identified resource control parameters crucial for distributing physical resources and characterizing contention for virtualized applications in a shared hosting environment. Second, we explored several modeling techniques and confirmed the suitability of two machine learning tools, Artificial Neural Network and Support Vector Machine, to accurately model the performance of virtualized applications. Moreover, we suggested and evaluated modeling optimizations necessary to improve prediction accuracy when using these modeling tools. Third, we presented an approach to optimal VM sizing by employing the performance models we created. Finally, we proposed a revenue-driven resource allocation algorithm which maximizes the SLA-generated revenue for a data center.