Hybrid address spaces: A methodology for implementing scalable high-level programming models on non-coherent many-core architectures

This paper introduces hybrid address spaces as a fundamental design methodology for implementing scalable runtime systems on many-core architectures without hardware support for cache coherence. We use hybrid address spaces for an implementation of MapReduce, a programming model for large-scale data processing, and the implementation of a remote memory access (RMA) model. Both implementations are available on the Intel SCC and are portable to similar architectures. We present the design and implementation of HyMR, a MapReduce runtime system whereby different stages and the synchronization operations between them alternate between a distributed memory address space and a shared memory address space, to improve performance and scalability. We compare HyMR to a reference implementation and we find that HyMR improves performance by a factor of 1.71× over a set of representative MapReduce benchmarks. We also compare HyMR with Phoenix++, a state-of-art implementation for systems with hardware-managed cache coherence in terms of scalability and sustained to peak data processing bandwidth, where HyMR demon- strates improvements of a factor of 3.1× and 3.2× respectively. We further evaluate our hybrid remote memory access (HyRMA) programming model and assess its performance to be superior of that of message passing.

Learning fluid prescription skills: why is it so challenging?

IIt is well recognised that medical students and junior doctors find fluid prescription a challenging topic. This study was designed to gain a greater understanding of the experiences that medical students face related to learning about fluid prescribing. Methods: A qualitative approach, using focus groups, was employed in this research. Final-year medical students in academic year 2011-12 at Queen's University Belfast were invited to participate during their 'Assistantship' placement in March 2012. Discussions in focus groups, consisting of between six and eight students, were recorded and transcribed verbatim. The research team, consisting of three separate investigators, conducted thematic analysis independently. A final consensus regarding emerging themes was reached by discussion within the whole research team. Medical students and junior doctors find fluid prescription a challenging topic Results: Five prominent themes emerged: 'Teaching experience: a disruptive variation'; 'Curricular disconnections'; 'The driving test: Theory-practice transformation'; 'Role modelling: which standard to aspire to?'; and finally 'Reconciling the perceived risk'. Discussion: This re search provided insights into medical students' opinions of the teaching practices and learning experiences related to fluid prescribing. The learning of prescribing skills is complex andcontextual. In the development of such skills, medical students are often exposed to conflicting educational experiences that challenge the novicelearner in making judgements on best prescribing practice. This study adds to the body of evidence that fluid prescription is a difficult topic, and has generated a number of multifaceted and strategic recommendations to potentially improve fluid prescription teaching.

A SPECIFICATION OF A COMPLEX PROGRAMMING LANGUAGE STATEMENT

A formal specification of a complex programming language statement is presented. The subject matter was selected as being typical of the kind confronting a small software house. It is shown that formal specification notations may be applied, with benefit, to 'messy' problems. Emphasis is placed upon producing a specification which is readable by, and useful to a reader not familiar with formal notations.

Design patterns percolating to parallel programming framework implementation

Structured parallel programming is recognised as a viable and effective means of tackling parallel programming problems. Recently, a set of simple and powerful parallel building blocks RISC pb²l) has been proposed to support modelling and implementation of parallel frameworks. In this work we demonstrate how that same parallel building block set may be used to model both general purpose parallel programming abstractions, not usually listed in classical skeleton sets, and more specialized domain specific parallel patterns. We show how an implementation of RISC pb² l can be realised via the FastFlow framework and present experimental evidence of the feasibility and efficiency of the approach.