Automatic memory management is crucial in implementation of runtime systems even though it induces a significant computational overhead. In this thesis I explore the use of statistical properties of the directed graph describing the set of live data to decide between garbage collection and heap expansion in a memory management algorithm combining the dynamic array represented heaps with a mark and sweep garbage collector to enhance its performance. The sampling method predicting the density and the distribution of useful data is implemented as a partial marking algorithm. The algorithm randomly marks the nodes of the directed graph representing the live data at different depths with a variable probability factor p. Using the information gathered by the partial marking algorithm in the current step and the knowledge gathered in the previous iterations, the proposed empirical formula predicts with reasonable accuracy the density of live nodes on the heap, to decide between garbage collection and heap expansion. The resulting heuristics are tested empirically and shown to improve overall execution performance significantly in the context of the Jinni Prolog compiler's runtime system.

There has recently been an increased emphasis on the importance of learning programming languages, not only in higher education but also in secondary schools. Students of a variety of departments such as physics, mathematics and engineering have also started learning programming languages as part of their academic courses. Assessment of students programming solutions is therefore important for developing their programming skills. Many Computer Based Assessment (CBA) systems utilise multiple-choice questions (MCQ) to evaluate students performance. However, MCQs lack the ability to comprehensively assess students knowledge. Thus, other forms of programming solutions are required to assess students knowledge. This research aims to develop a semi-automated assessment framework for novice programmers, utilising a computer to support the marking process. The research also focuses on ensuring the consistency of feedback. A novel marking process model is developed based on the semi-automated assessment approach which supports a new way of marking, termed segmented marking . A study is carried out to investigate and demonstrate the feasibility of the segmented marking technique. In addition, the new marking process model is developed based on the results of the feasibility study, and two novel marking process models are presented based on segmented marking, namely the full-marking and partial-marking process models. The Case-Based Reasoning (CBR) cycle is adopted in the marking process models in order to ensure the consistency of feedback. User interfaces of the prototype marking tools (full and partial) are designed and developed based on the marking process models and the user interface design requirements. The experimental results show that the full and partial marking techniques are feasible for use in formative assessment. Furthermore, the results also highlight that the tools are capable of providing consistent and personalised feedback and that they considerably reduce markers workload.