L'objectif de cette thèse est de présenter de nouveaux algorithmes de placement de machines virtuelles (VMs) à fin d’optimiser le coût et les émissions de carbone dans les Clouds distribués. La thèse se concentre d’abord sur la rentabilité des Clouds distribués, et développe ensuite les raisons d’optimiser les coûts ainsi que les émissions de carbone. La thèse comprend deux principales parties: la première propose, développe et évalue les algorithmes de placement statiques de VMs (où un premier placement d'une VM détient pendant toute la durée de vie de la VM). La deuxième partie propose des algorithmes de placement dynamiques de VMs où le placement initial de VM peut changer dynamiquement (par exemple, grâce à la migration de VMs et à leur consolidation). Cette thèse comprend cinq contributions. La première contribution est une étude de l'état de l'art sur la répartition des coûts et des émissions de carbone dans les environnements de clouds distribués. La deuxième contribution propose une méthode d'allocation des ressources, appelée NACER, pour les clouds distribués. L'objectif est de minimiser le coût de communication du réseau pour exécuter une tâche dans un cloud distribué. La troisième contribution propose une méthode de placement VM (appelée NACEV) pour les clouds distribués. NACEV est une version étendue de NACER. Tandis que NACER considère seulement le coût de communication parmi les DCs, NACEV optimise en même temps les coûts de communication et de calcul. Il propose également un algorithme de cartographie pour placer des machines virtuelles sur des machines physiques (PM). La quatrième contribution présente une méthode de placement VM efficace en termes de coûts et de carbone (appelée CACEV) pour les clouds distribués verts. CACEV est une version étendue de NACEV. En plus de la rentabilité, CACEV considère l'efficacité des émissions de carbone pour les clouds distribués. Pour obtenir une meilleure performance, la cinquième contribution propose une méthode dynamique de placement VM (D-CACEV) pour les clouds distribués. D-CACEV est une version étendue de notre travail précédent, CACEV, avec des chiffres supplémentaires, une description et également des mécanismes de migration de VM en direct. Nous montrons que notre mécanisme conjoint de réallocation-placement de VM peut constamment optimiser à la fois le coût et l'émission de carbone dans un cloud distribué / Virtual machine (VM) placement (i.e., resource allocation) method has a direct effect on both cost and carbon emission. Considering the geographic distribution of data centers (DCs), there are a variety of resources, energy prices and carbon emission rates to consider in a distributed cloud, which makes the placement of VMs for cost and carbon efficiency even more critical and complex than in centralized clouds. The goal of this thesis is to present new VM placement algorithms to optimize cost and carbon emission in a distributed cloud. It first focuses on cost efficiency in distributed clouds and, then, extends the goal to optimization of both cost and carbon emission at the same time. Thesis includes two main parts. The first part of thesis proposes, develops and evaluates static VM placement algorithms to reach the mentioned goal where an initial placement of a VM holds throughout the lifetime of the VM. The second part proposes dynamic VM placement algorithms where the initial placement of VMs is allowed to change (e.g., through VM migration and consolidation). The first contribution is a survey of the state of the art on cost and carbon emission resource allocation in distributed cloud environments. The second contribution targets the challenge of optimizing inter-DC communication cost for large-scale tasks and proposes a Network-Aware Cost-Efficient Resource allocation method, called NACER, for distributed clouds. The goal is to minimize the network communication cost of running a task in a distributed cloud by selecting the DCs to provision the VMs in such a way that the total network distance (hop count or any reasonable measure) among the selected DCs is minimized. The third contribution proposes a Network-Aware Cost Efficient VM Placement method (called NACEV) for Distributed Clouds. NACEV is an extended version of NACER. While NACER only considers inter-DC communication cost, NACEV optimizes both communication and computing cost at the same time and also proposes a mapping algorithm to place VMs on Physical Machines (PMs) inside of the selected DCs. NACEV also considers some aspects such as heterogeneity of VMs, PMs and switches, variety of energy prices, multiple paths between PMs, effects of workload on cost (energy consumption) of cloud devices (i.e., switches and PMs) and also heterogeneity of energy model of cloud elements. The forth contribution presents a Cost and Carbon Emission-Efficient VM Placement Method (called CACEV) for green distributed clouds. CACEV is an extended version of NACEV. In addition to cost efficiency, CACEV considers carbon emission efficiency and green distributed clouds. It is a VM placement algorithm for joint optimization of computing and network resources, which also considers price, location and carbon emission rate of resources. It also, unlike previous contributions of thesis, considers IaaS Service Level Agreement (SLA) violation in the system model. To get a better performance, the fifth contribution proposes a dynamic Cost and Carbon Emission-Efficient VM Placement method (D-CACEV) for green distributed clouds. D-CACEV is an extended version of our previous work, CACEV, with additional figures, description and also live VM migration mechanisms. We show that our joint VM placement-reallocation mechanism can constantly optimize both cost and carbon emission at the same time in a distributed cloud

To date, no observation-based proxy for climate change has been successful in quantifying the feedbacks between clouds and climate. The most promising, yet demanding, avenue to gain confi dence in cloud–climate feedback estimates is to utilize observations and large-eddy simulations (LES) or cloud-resolving modeling (CRM) to improve cloud process parameterizations in large-scale models. Sustained and improved satellite observations are essential to evaluate large-scale models. A reanalysis of numerical prediction models with assimilation of cloud, aerosol, and precipitation observations would provide a valuable dataset for examining cloud interactions. The link between climate modeling and numerical weather prediction (NWP) may be exploited by evaluating how accurate cloud characteristics are represented by the parameterization schemes in NWP models. A systematic simplifi cation of large-scale models is an important avenue to isolate key processes linked to cloud–climate feedbacks and would guide the formulation of testable hypotheses for fi eld studies. Analyses of observation-derived correlations between cloud and aerosol properties in combination with modeling studies may allow aerosol–cloud interactions to be detected and quantifi ed. Reliable representations of cloud dynamic and physical processes in large-scale models are a prerequisite to assess aerosol indirect effects on a large scale with confi dence. To include aerosol indirect effects in a consistent manner, we recommend that a “radiative fl ux perturbation” approach be considered as a complement to radiative forcing.

The Magellanic System at a distance of 50 kpc from the Milky Way (MW), is a prime target in the study of stellar populations, star formation histories and galactic dynamics in low metallicity environments. The Large Magellanic Cloud (LMC) and Small Magellanic Cloud (SMC) have been observed in great depth, however there has been somewhat less interest in the Magellanic Bridge connecting the two and only more recently has the interest surged in the Magellanic Stream, which trails the Clouds between them and the MW. The Magellanic Bridge has a known younger stellar population dating back to Irwin’s observations (Irwin et al., 1990), only more recently has an older population been confirmed in the Bridge by Bagheri et al. (2013) and No¨el et al. (2013), while the Magellanic Stream is known to contain gas only with no stellar component. The estimated ages of the Bridge and Stream are 200 Myr (Bekki, 2007) and 2 Gyr (Diaz and Bekki, 2012) respectively, with the postulated Bridge formation from a tidal interaction between the Clouds. The formation of the Stream is less well understood with different models using varying assumptions and parameters such as Besla et al. (2012) and Nidever et al. (2010), including possibilities that the Clouds were historically bound or un-bound, and that the MW may or may not have been involved in the Stream formation. The work in this thesis makes use of different methods of removing the Galactic foreground population in the direction of the Magellanic Bridge and Stream to create cleaned catalogues of these regions. Various methods of analysis are applied to the cleaned catalogues in this work to identify stellar populations in the Bridge and Stream and density variations in the Bridge, including the production of CMDs and two-colour diagrams, fitting isochrones to the observational data, creating stellar density maps and studying spatial variations. This work contains the first published confirmation that the Bridge contains an older population of stars from public catalogues, which is supported with observations of the older population in recent deeper surveys, and confirmed with spectroscopic follow up observations. The young population has ages within the age of the Bridge ( 200 − 500 Myr) and are likely to have formed in-situ, in regions of high density gas closest to the SMC. The number of young blue objects in the Bridge tiles is greater towards the SMC and decreases towards the LMC. Populations identified here reach ages up to 3 Gyr are likely to have been drawn into the Bridge from the Clouds at formation. The key results from this work are that an older Bridge stellar population has been identified and confirmed, indicating that stars as well as gas were drawn into the Bridge at its creation. The fact that the younger population has the highest density away from the main concentration of hydrogen show that the gas within the Bridge has been displaced by ram pressure, most likely due to the Clouds moving through the Galactic halo. Less concrete results in this work reveal a puzzling populetion of objects within the Magellanic Stream, which could be stellar in nature and with follow up work, could be the first observation of Stream objects. This work contributes to our understanding of the interaction between the LMC and SMC via the Stellar populations between them.

Massive star formation takes place in the dense cores of molecular clouds where the stars may be obscured at optical wavelengths. An excellent signpost of a massive young stellar object is the presence of an ultra-compact HII region (UCHii), which is a dense photo-ionised cocoon of gas surrounding the newly formed star. The aim of this project is to develop an assembly of numerical tools, caravela, that can simulate realistic data streams representing high-mass star forming regions in our Galaxy. The synthetic output consists in images and photometric point source catalogues, in the IRAS and Herschel wavebands. In an era when large observational surveys are increasingly important, this tool can produce simulated infrared point-source catalogues of high-mass star forming regions on a Galactic scale. The approach used is to construct a synthetic Galaxy of star-forming regions represented by SED templates. The star-forming regions are distributed randomly along a four spiral arm morphology, although a wide range of geometries can be used including rings and different numbers of spiral arms. The caravela code then observes the synthetic Galaxy to produce simulated images and point source catalogues with appropriate sensitivity and angular resolution. caravela was first used to model the simulated Galaxy by constraining the synthetic output to observations made by IRAS. This numerical tool will allow the user to infer physical properties of the Galactic population of high-mass star forming regions from such observations. Second, the selected model was again observed with caravela in Herschel mode. These are therefore predictive results for the future Herschel observations. A model with 4.0×104 compact proto-stars embedded in larger grey-body envelopes (with T = 40 K and linear size scale lIII = 5.0 × 106 AU) is the best-fit model to the IRAS observational data set studied. We found a level of contamination from low- and intermediate-mass objects of " 90%. The modelled data set resulting from the Herschel simulation resulted in the detection of approximately twice as many Herschel objects than IRAS, which is consistent, in a limited way, with the real observed companion clump fraction (CCF) of 0.90 ± 0.07 (Thompson et al., 2006) means that on average there were observed 2 sources per one IRAS source. Our caravela and the real observed CCF are therefore consistent. caravela was coupled with an independent diffuse emission model (Paladini et al., 2007) and the resulting analysis is presented as an interesting seed for the future.

A 'model-to-radiance' comparison of simulated brightness temperatures and radiances from the Hadley Centre Global Environmental Model 2 (HadGEM2-A) with longwave measurements from the High Resolution Infrared Radiation Sounder/4 (HIRS/4) and the Infrared Atmospheric Sounding Interfermeter (IASI) onboard the MetOp-A satellite is presented for all-sky and clear-sky global means. The fast Radiative Transfer model for TOVS 10 (RTTOV-10) is applied to HadGEM2 output to simulate observational-equivalent data. The results are compared with corresponding broadband analyses. A method is developed to extend hyperspectral IASI radiances to cover the whole outgoing terrestrial spectrum, in order to identify any compensating biases, and explore wavebands in the unobserved Far Infrared (FIR) region. For the all-sky HIRS analysis, the model overestimates brightness temperatures in the atmospheric window region with the greatest biases over areas associated with deep convective cloud. In contrast to many global climate models, much smaller clear-sky biases are found indicating that model clouds are the dominating source of error. Simulated values in upper atmospheric CO2 channels approximate observations better as a result of compensating cold biases at the poles and warm biases at lower latitudes, due to a poor representation of the Brewer Dobson circulation in the 38 level 'low-top' configuration of the model. Simulated all and clear-sky outgoing longwave radiation evaluated against the Clouds and the Earth's Radiant Energy System (CERES) and HIRS OLR products reveal good agreement, in part due to cancellation of positive and negative biases. Through physical arguments relating to the spectral energy balance within a cloud, it is suggested that broadband agreement could be the result of a balance between positive window biases and unseen negative biases originating from the water vapour rotational band in the FIR (not sampled by HIRS). Simple sensitivity tests show that dramatically altering existing cloud properties has little effect on the prominent window biases, however raising clouds a maximum of 5 atmospheric levels minimises the error in cloud contaminated channels, due to the introduction of spatially compensating errors. Sensitivities to the way ice clouds are parameterised in RTTOV-10 display a range of up to 2.5 K in window channels but absolute biases still exceed 3 K for all choices. Because of the lack of satellite based FIR observations due to a technological gap in the spectral region, an algorithm is created to 'fill in' the available data. Correlations between selected IASI channels and simulated unobserved wavelengths in the far infrared are used to estimate radiances between 25.25 - 644.75 cm-1 at 0.5 cm-1 intervals. The same method is used in the 2760 - 3000 cm-1 region. The spectrum is validated by comparing the Integrated Nadir Longwave Radiance (INLR) product (spanning the whole 25.25 - 3000 cm-1 range) with the corresponding broadband measurements from the Clouds and the Earth's Radiant Energy System (CERES) instrument on the Terra and Aqua satellites at simultaneous nadir overpasses, revealing mean differences of 0.3 Wm-2sr-1 (0.5% relative difference) lower for IASI relative to CERES and significantly lower biases in nighttime only scenes. Averaged global data over a single month produces mean differences of about 1 Wm-2sr-1 in both the all and the clear-sky (1.2% relative difference). The new high resolution spectrum is presented for global mean clear and total skies where the far infrared is shown to contribute 44% and 47% to the total OLR respectively, which is consistent with previous estimates. In terms of spectral cloud radiative forcing, the FIR contributes 19% and in some subtropical instances appears to be negative, results that would go un-observed with a traditional broadband analysis. The equivalent complete IASI OLR model product is simulated from GCM data using RTTOV-10. The same process of applying predictors to the satellite measurements is applied to the model simulated radiances, with appropriate modifications, to produce a directly comparable model product. Annual mean all-sky radiances are still greatly overestimated at all wavenumbers with a total radiance bias of 4.52 Wm-2 across the whole range. Compensating negative biases outside of the HIRS coverage that were hypothesised are absent, with the far infrared contributing to the overall bias rather than cancelling it. Equivalent clear-sky biases are much lower overall at 0.39 Wm-2, in part due to spectral and spatial cancellation of errors. A flux-to-flux comparison is enabled by estimating the spatial distribution of anisotropic factors, using collated HIRS OLR fluxes and IASI OLR radiances, which yields global mean model fluxes in excess of 12 Wm-2 higher than observations in the all-sky. The difference between this and the fluxes calculated using the climate model's broadband radiation code (Edward-Slingo) are around 10 Wm-2 which is outside the range of uncertainty in the method used to estimate the flux. However, it is discussed that tuning of the climate model's broadband code to known flux values is a required practice to ensure global energy budgets balance but can produce inaccurate parameterised variables. An equivalent analysis adjusting the ice cloud parametrisation to reflect the radiances that have the biggest differences to the original configuration selected showed a bias reduction of 4.5 Wm-2, which is still not enough to completely explain its size, suggesting the existence of residual cloud problems. Finally, it is suggested that the way forward in separating and constraining cloud errors, in both radiative transfer codes, is a rigorous process of testing them with observation cloud properties and reanalysis data as inputs.

In this paper, we introduce the Local Volume TiNy Titans sample (LV-TNT), which is a part of a larger body of work on interacting dwarf galaxies: TNT . This LV-TNT sample consists of 10 dwarf galaxy pairs in the Local Universe (< 30 Mpc from Milky Way), which span mass ratios of M-*,M- 1/M-*,M- 2 < 20, projected separations < 100 kpc, and pair member masses of log(M-*/M-aS (TM)) < 9.9. All 10 LV-TNT pairs have resolved synthesis maps of their neutral hydrogen, are located in a range of environments and captured at various interaction stages. This enables us to do a comparative study of the diffuse gas in dwarf-dwarf interactions and disentangle the gas lost due to interactions with haloes of massive galaxies, from the gas lost due to mutual interaction between the dwarfs. We find that the neutral gas is extended in the interacting pairs when compared to non-paired analogues, indicating that gas is tidally pre-processed. Additionally, we find that the environment can shape the H i distributions in the form of trailing tails and that the gas is not unbound and lost to the surroundings unless the dwarf pair is residing near a massive galaxy. We conclude that a nearby, massive host galaxy is what ultimately prevents the gas from being re-accreted. Dwarf-dwarf interactions thus represent an important part of the baryon cycle of low-mass galaxies, enabling the 'parking' of gas at large distances to serve as a continual gas supply channel until accretion by a more massive host.

Stars are born from dense cores of gas within molecular clouds. The exact nature of the connection between these gas cores and the stars they form is an important issue in the field of star formation. In this thesis I use numerical simulations of molecular clouds to trace the evolution of cores into stars. The CLUMPFIND method, commonly used to identify gas structures is tested. I find that the core boundaries it yields are unreliable, but in spite of this, the same profile is universally found for the mass function. To facilitate a more robust definition of a core, a modified clumpfind algorithm which uses gravitational potential instead of density is introduced. This allows the earliest fragmentation in a simulated molecular cloud to be identified. The first bound cores have a mass function that closely resembles the stellar IMF, but there is a poor correspondence between individual core masses and the stellar masses formed from them. From this, it is postulated that environmental factors play a significant part in a core’s evolution. This is particularly true for massive stars, as massive cores are prone to further fragmentation. In these simulations, massive stars are formed simultaneously with stellar clusters, and thus the evolution of one can affect the other. In particular, the global collapse of the forming cluster aids accretion by the precursors of the massive stars. By tracing the evolution of the massive stars, I find that most of the material accreted by them comes from diffuse gas, rather than from a well-defined stellar core.

Various physical mechanisms that affect radiowave propagation at millimetre wavelengths are considered. Current modelling weaknesses are highlighted and new improved models or more appropriate modelling approaches are suggested. Interference and resonance phenomena in the scattering of spherical ice and water particles are reviewed. The long standing problem of the numerous resonances observed in the scattering diagrams of dielectric spheres is answered. The spatial structure and the physical characteristics of non-precipitable ice and water clouds are reviewed. Extinction and back scattering calculations for a wide variety of cloud models over the entire millimetre frequency spectrum are given. Multiple scattering and the effects of super-large drops in clouds are also dealt with. The potential of a spaceborne instrument in deducing information about the vertical structure of various cloud types is examined. Attenuation and reflectivity profiles resulting from various cloud types are calculated for a nadir pointing fixed beam millimetre wave radar operating at 94 GHz. The physics and application of the equation of radiative transfer to millimetre wave propagation in the earth's atmosphere are given and also is the solution of this equation for a typical millimetre wave remote sensing application. The theory of gaseous absorption at millimetre wavelengths is presented and an improved modelling approach is proposed for the calculation of the absorption and dispersion spectra of atmospheric gases. The effects of trace gases on communication systems operating at high altitudes are for the first time reported. Finally the use of the 60 GHz oxygen absorption band for top-side air traffic control/navigation and broadband transmission purposes is studied.

The interstellar medium (ISM) of our Galaxy, and of others, is pervaded by ultra low-density gas and dust, as well as magnetic fields. Embedded magnetic fields have been known to play an important role in the structure and dynamics of the ISM. However, the ability to accurately quantify these fields has plagued astronomers for many decades. Unfortunately, the experimental techniques for measuring the strength and direction of magnetic fields are few, and they are observationally challenging. The only direct method of measuring the magnetic field is through the Zeeman effect. The goal of this dissertation is to expand upon the current observational studies and understanding of the effects of interstellar magnetic fields across various regions of the Galaxy. Zeeman effect observations of magnetic fields in two dynamically diverse environments in the Milky Way are presented: (1) An OH and HI absorption line study of envelopes of molecular clouds distributed throughout the Galaxy, and (2) A study of OH absorption lines toward the Galactic center region in the vicinity of the supermassive black hole Sgr A*. We have executed the first systematic observational survey designed to determine the role of magnetic fields in the inter-core regions of molecular clouds. Observations of extragalactic continuum sources that lie along the line-of-sight passing through Galactic molecular clouds were studied using the Arecibo telescope. OH Zeeman effect observations were combined with estimates of column density to allow for computation of the mass-to-flux ratio, a measurement of the gravitational to magnetic energies within a cloud. We find that molecular clouds are slightly subcritical overall. However, individual measurements yield the first evidence for magnetically subcritical molecular gas. Jansky VLA observations of 18 cm OH absorption lines were used to determine the strength of the line-of-sight magnetic field in the Galactic center region. This study yields no clear detections of the magnetic field and results that differ from a similar study by Killeen, Lo, & Crutcher (1992). Our results suggest magnetic fields no more than a few microgauss in strength.

Developing and maintaining software is a difficult task, and finding effective methods of understanding software is more necessary now than ever with the last few decades seeing a dramatic climb in the scale of software. Appropriate visualisations may enable greater understanding of the datasets we deal with in software engineering. As an aid for sense-making, visualisation is widely used in daily life (through graphics such as weather maps and road signs), as well as in other research domains, and is thought to be exceedingly beneficial. Unfortunately, there has not been widespread use of the multitude of techniques which have proposed for the software engineering domain. Tag clouds are a simple, text-based visualisation commonly found on the internet. Typically, implementations of tag clouds have not included rich interactive features which are necessary for data exploration. In this thesis, I introduce design considerations and a task set for enabling interaction in a tag cloud visualisation system. These considerations are based on an analysis of challenges in visualising software engineering data, and the perceptive influences of visual properties available in tag clouds. The design and implementation of interactive system Taggle based on these considerations is also presented, along with its broad-based evaluation. Evaluation approaches were informed by a systematic mapping study of previous tag cloud evaluation, providing an overview of existing research in the domain. The design of Taggle was improved following a heuristic evaluation by domain experts. Subsequent evaluations were divided into two parts - experiments focused on the tag cloud visualisation technique itself, and a task-based approach focused on the whole interactive system. As evidenced in the series of evaluative studies, the enhanced tag cloud features incorporated into Taggle enabled faster visual search response time, and the system could be used with minimal training to discover relevant information about an unknown software engineering dataset.