published_or_final_version / Computer Science / Master / Master of Philosophy

This dissertation demonstrates new methods for increasing the performance and scalability of high performance networking technologies for use in clustered computing systems, concentrating on Ethernet/High-Speed networking convergence. The motivation behind the improvement of high performance networking technologies and their importance to the viability of modern data centers is discussed first. It then introduces the concepts of high performance networking in a commercial data center context as well as high performance computing (HPC) and describes some of the most important challenges facing such networks in the future. It reviews current relevant literature and discusses problems that are not yet solved. Through a study of existing high performance networks, the most promising features for future networks are identified. Sockets Direct Protocol (SDP) is shown to have unexpected performance issues for commercial applications, due to inefficiencies in handling large numbers of simultaneous connections. The first SDP over eXtended Reliable Connections implementation is developed to reduce connection management overhead, demonstrating that performance issues are related to protocol overhead at the SDP level. Datagram offloading for IP over InfiniBand (IPoIB) is found to work well. In the first work of its kind, hybrid high-speed/Ethernet networks are shown to resolve the issues of SDP underperformance and demonstrate the potential for hybrid high-speed networking local area Remote Direct Memory Access (RDMA) technologies and Ethernet wide area networking for data centers. Given the promising results from these studies, a set of solutions to enhance performance at the local and wide area network level for Ethernet is introduced, providing a scalable, connectionless, socket-compatible, fully RDMA-capable networking technology, datagram-iWARP. A novel method of performing RDMA Write operations (called RDMA Write-Record) and RDMA Read over unreliable datagrams over Ethernet is designed, implemented and tested. It shows its applicability in scientific and commercial application spaces and is applicable to other verbs-based networking interfaces such as InfiniBand. The newly proposed RDMA methods, both for send/recv and RDMA Write-Record, are supplemented with interfaces for both socket-based applications and Message Passing Interface (MPI) applications. An MPI implementation is adapted to support datagram-iWARP. Both scalability and performance improvements are demonstrated for HPC and commercial applications. / Thesis (Ph.D, Electrical & Computer Engineering) -- Queen's University, 2012-09-25 09:43:55.262

Selenonium cations are electron deficient species in which the central selenium atom is bonded to three carbon chains (aryl or alkyl). Trimethylselenonium iodide was synthesized by reaction of methyllithium with metallic selenium to produce methylselenolithium which was, in turn, reacted with the appropriate alkylbromide. The selenide thus formed was further methylated at the selenium atom with methyl iodide in methanol in the presence of sodium tetraphenylborate. After several recrystallizations the selenonium analytes were characterized by AAS, FT-IR, $ sp1$H-NMR, $ sp{13}$C-NMR, FAB-MS and LAMMA spectroscopic techniques and used as standards for analytical methods development. / The analysis was performed by high performance liquid chromatography with atomic absorption detection. The chromatography on a cynopropyl silica bonded phase was optimized for mobile phase composition by response surface analysis. The resulting surface response plots permitted a differentiation between the mechanisms of action of two mobile phase modifiers: triethylamine and trimethylsulfonium iodide. The improvement in chromatographic efficiency resulted in two to three fold decrease in the limit of detection. An extraction procedure with liquefied phenol was evaluated for the determination, by HPLC-AAS, of traces of selenonium cations in biological samples. The advantages and shortcomings of the HPLC-THG-AAS approach are discussed.

This research was undertaken to study the effects of high performance concrete (HPC) mix design modifications on the propensity of early-age cracking. Seven mixtures were tested: one 35 MPa conventional concrete (CC) mixture made with ordinary Portland cement with blended slag; one typical 50 MPa HPC mixture containing slag and silica fume; and five modified HPC mixtures using extra set-retarder, increased slag replacement, shrinkage-reducing admixture (SRA), pre-saturated lightweight aggregate (LWA), and decreased cement paste content to improve thermal and/or shrinkage properties. The mixtures were tested for durability, mechanical, thermal, and shrinkage properties. All modified HPC mixtures showed reduced shrinkage relative to the HPC control mixture, and the most shrinkage mitigation was observed in the mixture containing LWA. While SRA reduced restrained shrinkage in HPC to the magnitude of CC, it provided very low rapid chloride penetrability, and using LWA in HPC resulted in significant restrained shrinkage reduction compared to CC.

This research was undertaken to study the effects of high performance concrete (HPC) mix design modifications on the propensity of early-age cracking. Seven mixtures were tested: one 35 MPa conventional concrete (CC) mixture made with ordinary Portland cement with blended slag; one typical 50 MPa HPC mixture containing slag and silica fume; and five modified HPC mixtures using extra set-retarder, increased slag replacement, shrinkage-reducing admixture (SRA), pre-saturated lightweight aggregate (LWA), and decreased cement paste content to improve thermal and/or shrinkage properties. The mixtures were tested for durability, mechanical, thermal, and shrinkage properties. All modified HPC mixtures showed reduced shrinkage relative to the HPC control mixture, and the most shrinkage mitigation was observed in the mixture containing LWA. While SRA reduced restrained shrinkage in HPC to the magnitude of CC, it provided very low rapid chloride penetrability, and using LWA in HPC resulted in significant restrained shrinkage reduction compared to CC.

No description available.

Amadori rearrangement product (ARP) of tryptophan with glucose was synthesized according to a published procedure, and its decomposition was studied at two different concentrations and at two temperatures, 110$ sp circ$C and 140$ sp circ$C over a period of 6 hrs. A RP-HPLC system was developed to separate and quantitate the decomposition products of the ARP. The results indicated that, the ARP can decompose to form the following products, hydroxymethylfurfural (HMF), maltol, indole, $ beta$-carbolines (norharman and harman) and tryptophan. Further, using the same analytical method, the following systems were also analyzed for the presence of the above mentioned products (a) D-glucose and D-mannose with tryptophan; (b) D-glucose; and (c) tryptophan. In addition, rate constants and activation energies for the decomposition and formation reaction were calculated. Plausible mechanisms for the formation of the decomposition products are suggested.

In cellular studies of patients with lipid related disorders such as mammary cancers, leukemia, and artheroscierosis, separation of molecular species of oxidized phosphatidylcholine (PC) can be an important assistance in research or diagnosis. Goals of this project were to optimize separation of oxidized and unoxidized PC molecular species in a single HPLC chromatogram followed by in line identification of hydroperoxides. Oxidized egg PC's were produced using UV light exposure in air. Separations were performed on an Ultrasphere ODS column and an Asahipak ODPVA column using a Waters 2695 system with photodiode array. The ODPVA column routinely gave 10 times larger plate numbers. Various mobile phase mixtures (methanol, acetonitrile, water) and gradients were tested. The optimum gradient on our system is (1) 5 minutes, 47 % methanol/40 % acetonitrile/13 water in a linear gradient to (2) 17 minutes, 49 % methanol/40 % acetonitrile/11 % water to (3) 18 minutes, 29 % methanol/60 % acetonitrile/11 % water linearly to (4) 50 minutes, 31 % methanol/60 % acetonitrile/9 % water continued isocratically to 110 minutes. Oxidized hydroperoxides were detected by fluorescence using a post column reaction with diphenyl-1 pyrenylphosphine (DPPP). Both iron (III) and pyridine were tested as catalysts for this reaction. / Department of Chemistry

Using archival analysis and interviews this study examined the career pathways, learning experiences, and athletic experiences of 11 high-performance head hockey coaches who played goal in ice hockey. Guided by the learning situations discussed in Wright, Trudel, and Culver (2007) the interviews revealed four important learning experiences common to all the coaches: coach interactions, books and videotapes, coach clinics and academic education, and experiences related to playing and coaching. Like Werthner and Trudel (2009) the results indicated that certain similarities aside, each coach’s career pathway is idiosyncratic with elite athletic experience being an important, but not imperative, aspect of high-performance coaches’ career development. The findings provide insight into how these individuals acquired their coaching knowledge and provide a more complete picture of the developmental pathways associated with becoming a high-performance head hockey coach.

Thesis (Ph.D.)--University of Texas at Dallas, 2007. / Includes vita. Includes bibliographical references (leaves 94-103)