Emulation is the practice of simulating one computer system on another. There are many methods of implementing an emulator. They exist on a performance continuum from simple interpretation to dynamic binary translation extended with various optimizations. Optimizations are diverse, including just in time compilation, large translation units, shadow stack, register mapping and many more. The goal of this thesis is to develop a high performance, portable emulator for the ARM v4 architecture without requiring substantial code analysis. This thesis describes the implementation of a dynamic binary translator translating to an intermediate language targeting a virtual machine. Targeting a virtual machine ensures that the emulator is portable. Optimizations implemented include forming large translation units and branch straightening in hot regions. The particular combination of translating to intermediate form for a virtual machine, and creating large translation units from hot regions does not seem to appear in the literature. The performance of the described dynamic binary translator exceeds the performance of an interpreter on the same platform by an order of magnitude. Code analysis was only used to straighten branches in hot regions. While many popular dynamic binary translation optimizations are not readily applicable when using a virtual machine target, the performance achieved shows that using virtual machine as translation target is viable method of implementing dynamic binary translator. / Graduate

It has become increasingly clear that many binary systems will pass through a common envelope stage at some point during their evolution. For short period systems composed of main-sequence 0 and early B stars this stage will probably occur for the first time towards the end of hydrogen-core burning in the primary component (case A evolution) rather than during the transition to the giant stage (case B evolution). If masses, radii, luminosities, temperatures and orbital parameters were well determined for a good sample of those systems, it could be established whether the individual components were so close that case A evolution was inevitable or whether the primary component had enough room to complete its main-sequence phase before reaching its Roche. limit and hence case B evolution. The latter mode has been studied extensively (both conservatively and non-conservatively) whereas the reception of matter by the secondary component in the rapid mass-transfer phase of case A evolution has only very recently been investigated. It is still one of the principal problems impeding further progress on this scenario. To resolve this situation and provide observational material with which to compare these theoretical models, an observing program was established to study systems of spectral type earlier than B5 and of orbital period of less than 1.8 in both the Northern and Southern Hemispheres. Light curves were obtained at St Andrews using the newly-refurbished Twin Photometric Telescope and analysed using software developed specifically for this instrument. Further spectroscopic and photometric data were obtained at La Palma SAAO, Sierra Nevada and Boyden. Analyses of these spectroscopic and photometric observations have provided the necessary physical parameters to determine the evolutionary status of these systems. The systems observed were AH Cephei and V1182 Aquilae which are shown to be detached systems, TT Aurigae, SX Aurigae and AI Crucis which are all semi-detached systems and V701 Scorpii and HZ Pyxidis which are contact systems. Accurately-determined parameters of 14 stars have been found, including four 0 stars in detached systems. Therefore-the number of stars with well-determined masses of greater than 30M has been increased by 25%. It is clear from this study that case A mass transfer will play and has played an important role in the evolution of five out of the seven systems. It is debatable whether or not the contact systems have passed through a mass transfer phase, particularly RZ Pyx. The evolutionary history of this system is of particular interest, especially if this binary was in a marginal contact configuration when it arrived on the main sequence. Attempts have been made to look for intrinsic variability in these systems but no periodic variation has been found in any of them. If such a phenomenon exists in one of the components of the binaries in the sample then it must have an amplitude of less than 0.01. The comparison of the physical parameters of 67 stars compiled by the author from this work and from published data with theoretical zero-age and terminal-age main sequences shows that traditional modelling of semi-convection without mass loss is not adequate. Convective overshooting and mass loss play a very important role in the evolution of massive close binary systems of short period.

Orbits of spectroscopic binary systems have been studied for more than a century. Over three thousand orbits of spectroscopic binary systems have been derived. These orbits are based on the radial velocities measured from the spectra recorded by a photographic plate to a high precision spectrum observed from a modern spectrograph. In many cases, the shape of the orbit was assumed to be circular, of hence the eccentricity is zero. This assumption is based on the fact that a small eccentricity (e < 0.1) measured from the observed data might be a result from the error of observations or from the intrinsic variation of a spectroscopic binary system. Sixteen southern spectroscopic binary systems, including twelve single-lined binaries and four double-lined binaries, were selected to study in this research program. These systems were assumed to have circular orbits or have very nearly circular orbits (e < 0.1) from their previous published solutions. The HERCULES spectrograph was used in conjunction with the 1-m McLellan telescope at Mt John University Observatory to collect the spectra of these systems. The observations, taken from October 2004 to August 2007, comprised about 2000 high-resolution spectra of spectroscopic binary systems and standard radial-velocity stars. Radial velocities of spectroscopic binary systems were measured from these spectra and orbital solutions of the systems were derived from these radial velocities. It was found that from HERCULES data, we are able to achieve high-precision orbital solutions of all the systems studied. The best-fit solutions can be improved as much as 70 times from the literature’s orbital solutions. It has been found that the precision of a system depends on the rotational velocities of the components as well as the level of their chromospheric activity. We are able to confirm the eccentricity in the orbit of only one of the selected spectroscopic binary systems, HD194215. Its eccentricity is 0.123 29 ± 0.000 78. The small eccentricities of other systems are not confirmed. There are four systems; HD22905, HD38099, HD85622 and HD197649, that have circular orbital solutions from the large errors in their measured eccentricities. Two systems, HD77258 and HD124425, have too small eccentricities, e = 0.000 85±0.000 19 and 0.002 60 ± 0.000 99 to be acceptable. An intrinsic variation is a presumed cause of the spurious eccentricities derived from the data of the other eight systems. Photometric data from Mt John University Observatory service photometry program, as well as the photometric data from the Hipparcos satellite and information of these systems from the literature, using various methods and instruments, give a wider view on the systems’ behaviour. It is possible that the spurious eccentricities derived for these systems result from the eclipsing behaviour of a system (HD50337), or from the nature of the components, such as, the distortion of their shape (HD352 and HD136905), their chromospheric activity (HD9053, HD3405, HD77137, HD101379 and HD155555), or stellar pulsation (HD30021). Models of the active chromosphere system, HD101379, have been simulated. An analysis of synthetic radial velocity data shows that spots on the star’s photosphere can cause a spurious eccentricity. The values of the spurious eccentricity and the longitude of periastron are dependent on the spot size, the spot temperature, and the position of the spots.

Inaug.-dis.--Königsberg.

Thesis (Ph. D.)--University of Wisconsin--Madison, 1904. / Reprinted from Popular astronomy, vol. XIII, nos. 5 and 6, May and June-July, 1905.

Thesis (M. Phil.)--University of Hong Kong, 1996. / Includes bibliographical references (leaf 198-214).

Thesis (M. Phil.)--University of Hong Kong, 1994. / Includes bibliographical references (leaves 134-135).

Thesis (Ph. D.)--Johns Hopkins University, 1913. / Biographical sketch. "Reprinted from American journal of mathematics, vol. XXXVI, no. 3, July, 1914."

Thesis (M.Phil.)--University of Hong Kong, 1996. / Includes bibliographical references (leaf 198-214). Also available in print.

Thesis (Ph. D.)--University of Wisconsin--Madison, 1982. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 261-269).