A User-Centric Perspective on Parallel Programming with Focus on OpenMP

The process of developing software that takes advantage of multiple processors is commonly referred to as parallel programming. For various reasons, this process is much harder than the sequential case. For decades, parallel programming has been a problem for a small niche only: engineers working on parallelizing mostly numerical applications in High Performance Computing. This has changed with the advent of multi-core processors in mainstream computer architectures. Parallel programming in our days becomes a problem for a much larger group of developers. The main objective of this thesis was to find ways to make parallel programming easier for them. Different aims were identified in order to reach the objective: research the state of the art of parallel programming today, improve the education of software developers about the topic, and provide programmers with powerful abstractions to make their work easier. To reach these aims, several key steps were taken. To start with, a survey was conducted among parallel programmers to find out about the state of the art. More than 250 people participated, yielding results about the parallel programming systems and languages in use, as well as about common problems with these systems. Furthermore, a study was conducted in university classes on parallel programming. It resulted in a list of frequently made mistakes that were analyzed and used to create a programmers' checklist to avoid them in the future. For programmers' education, an online resource was setup to collect experiences and knowledge in the field of parallel programming - called the Parawiki. Another key step in this direction was the creation of the Thinking Parallel weblog, where more than 50.000 readers to date have read essays on the topic. For the third aim (powerful abstractions), it was decided to concentrate on one parallel programming system: OpenMP. Its ease of use and high level of abstraction were the most important reasons for this decision. Two different research directions were pursued. The first one resulted in a parallel library called AthenaMP. It contains so-called generic components, derived from design patterns for parallel programming. These include functionality to enhance the locks provided by OpenMP, to perform operations on large amounts of data (data-parallel programming), and to enable the implementation of irregular algorithms using task pools. AthenaMP itself serves a triple role: the components are well-documented and can be used directly in programs, it enables developers to study the source code and learn from it, and it is possible for compiler writers to use it as a testing ground for their OpenMP compilers. The second research direction was targeted at changing the OpenMP specification to make the system more powerful. The main contributions here were a proposal to enable thread-cancellation and a proposal to avoid busy waiting. Both were implemented in a research compiler, shown to be useful in example applications, and proposed to the OpenMP Language Committee.

Effects of grassland renovation on carbon concentrations and aggregate distribution in temperate grassland soils

The effects of continuous tillage on the distribution of soil organic matter (SOM) and aggregates have been well studied for arable soils. However, less is known about the effects of sporadic tillage on SOM and aggregate dynamics in grassland soils. The objectives of the present thesis were (I) to study the longer-term effects of sporadic tillage of grassland on organic carbon (Corg) stocks and the distribution of aggregates and SOM, (II) to investigate the combined effects of sporadic tillage and fertilization on carbon and nitrogen dynamics in grassland soils, and (III) to study the temporal dynamics of Corg stocks, aggregate distribution and microbial biomass in grassland soils. Soil samples were taken in three soil depths (0 – 10 cm; 10 – 25 cm; 25 – 40 cm) from a field trial with loamy sandy soils (Cambisols, Eutric Luvisols, Stagnosols, Anthrosols) north of Kiel, Germany. For Objective I we have sampled soil two and five years after one or two tillage operation(s). Treatments consisted of (i) permanent grassland, (ii) tillage of grassland followed by a re-establishment of grassland and (iii) tillage of grassland followed by a re-establishment of grassland with one season of winter wheat in between. The tillage in grassland led to a reduction in Corg stocks, large macroaggregates (>2000 µm) and SOM in the top 10 cm soil depth. These findings were still significant two years after tillage; however, five years after tillage no longer present. Regarding the soil profile (0 – 40 cm) no significant differences in the mentioned parameters between the tilled plots and the permanent grassland existed. A second tillage event and the insertion of one season of winter wheat did not lead to any further effects on Corg stocks as well as aggregate and SOM concentrations in comparison with a single tillage event in these grassland soils. Treatments adapted for Objective II included (i) long-term grassland and (ii) tillage of grassland followed by a re-establishment of grassland with one season of winter wheat in between. The plots were split and received either 240 kg N ha-1 year-1 in the form of cattle slurry or no cattle slurry application. The application of slurry within a period of four years had no effects on the Corg and total nitrogen stocks or the aggregate distribution, but led to a reduction of free and not physically protected SOM. However, the application of cattle slurry and the grassland renovation seems to change the plant species composition and therefore generalizations on the direct effects are not yet possible. For studying Objective III a further field trial was initiated in September 2010. Soil samples were taken six times within one year (from October 2010 to October 2011) (i) after the conversion from arable land into grassland, (ii) after the tillage of grassland followed by a re-establishment of grassland and (iii) in a permanent grassland. We found an increase in the microbial and fungal biomass after the conversion of arable land into grassland, but no effect on aggregate distribution and Corg stocks. A one-time tillage operation in grassland led to a reduction in large macroaggregates and Corg stocks in the top 10 cm soil depth with no effect on the sampled soil profile. However, we found large variations in the fungal biomass and aggregate distribution within one year in the permanent grassland, presumably caused by environmental factors. Overall, our results suggest that a single tillage operation in grassland soils markedly decreased the concentrations of Corg, larger aggregates and SOM. However, this does not result in long-lasting effects on the above mentioned parameters. The application of slurry cannot compensate the negative effects of a tillage event on aggregate concentrations or Corg stocks. However, while the Corg concentration is not subject to fluctuations within a year, there are large variations of the aggregate distribution even in a permanent grassland soil. Therefore conclusions of results from a single sampling time should be handled with care.

Optimization in the Technological Steps for the Fabrication of Large Area Micromirror Arrays

Micromirror arrays are a very strong candidate for future energy saving applications. Within this work, the fabrication process for these micromirror arrays has been optimized and some steps for the large area fabrication of micromirror modules were performed. At first the surface roughness of the insulation layer of silicon dioxide (SiO2) was investigated. This SiO2 thin layer was deposited on three different type of substrates i.e. silicon, glass and Polyethylene Naphthalate (PEN) substrates. The deposition techniques which has been used are Plasma Enhanced Chemical Vapor Deposition (PECVD), Physical Vapor Deposition (PVD) and Ion Beam Sputter Deposition (IBSD). The thickness of the SiO2 thin layer was kept constant at 150nm for each deposition process. The surface roughness was measured by Stylus Profilometry and Atomic Force Microscopy (AFM). It was found that the layer which was deposited by IBSD has got the minimum surface roughness value and the layer which was deposited by PECVD process has the highest surface roughness value. During the same investigation, the substrate temperature of PECVD was varied from 80° C to 300° C with the step size of 40° C and it was found that the surface roughness keeps on increasing as the substrate holder temperature increases in the PECVD process. A new insulation layer system was proposed to minimize the dielectric breakdown effect in insulation layer for micromirror arrays. The conventional bilayer system was replaced by five layer system but the total thickness of insulation layer remains the same. It was found that during the actuation of micromirror arrays structure, the dielectric breakdown effect was reduced considerably as compared to the bilayer system. In the second step the fabrication process of the micromirror arrays was successfully adapted and transferred from glass substrates to the flexible PEN substrates by optimizing the conventional process recipe. In the last section, a large module of micromirror arrays was fabricated by electrically interconnecting four 10cm×10cm micromirror modules on a glass pane having dimensions of 21cm×21cm.