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.

Productivity growth and the structure of production

In this study, interactions between potential hierarchical value chains existing in the production structure and industry-wise productivity growths are sought. We applied generalized Chenery-Watanabe heuristics for matrix linearity maximization to triangulate the input-output incidence matrix for both Japan and the Republic of Korea, finding the potential directed flow of values spanning the industrial sectors of the basic (disaggregated) industry classifications for both countries. Sector specific productivity growths were measured by way of the Trönquvist index, using the 2000-2005 linked input-output tables for both Japan and Korea.

Age and growth of the bigeye thresher shark, Alopias superciliosus, from the pelagic longline fisheries in the tropical northeastern Atlantic Ocean, determined by vertebral band counts

The bigeye thresher, Alopias supercilious, is commonly caught as bycatch in pelagic longline fisheries targeting swordfish. Little information is yet available on the biology of this species, however. As part of an ongoing study, observers sent aboard fishing vessels have been collecting set of information that includes samples of vertebrae, with the aim of investigating age and growth of A. supercilious. A total of 117 specimens were sampled between September 2008 and October 2009 in the tropical northeastern Atlantic, with specimens ranging from 101 to 242 cm fork length (FL) (176 to 407 cm total length). The A. supercilious vertebrae were generally difficult to read, mainly because they were poorly calcified, which is typical of Lamniformes sharks. Preliminary trials were carried out to determine the most efficient band enhancement technique for this species, in which crystal violet section staining was found to be the best methodology. Estimated ages in this sample ranged from 2 to 22 years for females and 1 to 17 years for males. A version of the von Bertalanffy growth model (VBGF) re-parameterised to estimate L(0), and a modified VBGF using a fixed L(0) were fitted to the data. The Akaike information criterion (AIC) was used to compare these models. The VBGF produced the best results, with the following parameters: L(inf) = 293 cm FL, k = 0.06 y(-1) and L(0) = 111 cm FL for females; L(inf) = 206 cm FL, k = 0.18 y(-1) and L(0) = 93 cm FL for males. The estimated growth coefficients confirm that A. supercilious is a slow-growing species, highlighting its vulnerability to fishing pressure. It is therefore urgent to carry out more biological research to inform fishery managers more adequately and address conservation issues.

Growth of boehmite nanofibers by assembling nanoparticles with surfactant micelles

It is known that boehmite (AlOOH) nanofibers formed in the presence of nonionic poly(ethylene oxide) (PEO) surfactant at 373 K. A novel approach is proposed in this study for the growth of the boehmite nanofibers: when fresh aluminum hydrate precipitate was added at regular interval to initial mixture of boehmite and PEO surfactant at 373 K, the nanofibers grow from 40 to 50 nm long to over 100 nm. It is believed that the surfactant micelles play an important role in the nanofiber growth: directing the assembly of aluminum hydrate particles through hydrogen bonding with the hydroxyls on the surface of aluminum hydrate particles. Meanwhile a gradual improvement in the crystallinity of the fibers during growth is observed and attributed to the Ostwald ripening process. This approach allows us to precisely control the size and morphology of boehmite nanofibers using soft chemical methods and could be useful for low temperature, aqueous syntheses of other oxide nanomaterials with tailorable structural specificity such as size, dimension and morphology.