A Unified Scheduler for Recursive and Task Dataflow Parallelism

Task dataflow languages simplify the specification of parallel programs by dynamically detecting and enforcing dependencies between tasks. These languages are, however, often restricted to a single level of parallelism. This language design is reflected in the runtime system, where a master thread explicitly generates a task graph and worker threads execute ready tasks and wake-up their dependents. Such an approach is incompatible with state-of-the-art schedulers such as the Cilk scheduler, that minimize the creation of idle tasks (work-first principle) and place all task creation and scheduling off the critical path. This paper proposes an extension to the Cilk scheduler in order to reconcile task dependencies with the work-first principle. We discuss the impact of task dependencies on the properties of the Cilk scheduler. Furthermore, we propose a low-overhead ticket-based technique for dependency tracking and enforcement at the object level. Our scheduler also supports renaming of objects in order to increase task-level parallelism. Renaming is implemented using versioned objects, a new type of hyper object. Experimental evaluation shows that the unified scheduler is as efficient as the Cilk scheduler when tasks have no dependencies. Moreover, the unified scheduler is more efficient than SMPSS, a particular implementation of a task dataflow language.

Separate Density and Viscosity Determination of Room Temperature Ionic Liquids using Dual Quartz Crystal Microbalances

The drive towards cleaner industrial processes has led to the development of room temperature ionic liquids (RTIL) as environmentally friendly solvents. They comprise solely of ions which are liquid at room temperature and with over one million simple RTIL alone it is important to characterize their physical properties using minimal sample volumes. Here we present a dual Quartz Crystal Microbalance (QCM) which allows separate determination of viscosity and density using a total sample volume of only 240 mu L. Liquid traps were fabricated on the sensing area of one QCM using SU-8 10 polymer with a second QCM having a flat surface. Changes in the resonant frequencies were used to extract separate values for viscosity and density. Measurements of a range of pure RTIL with minimal water content have been made on five different trap designs. The best agreement with measurements from the larger volume techniques was obtained for trap widths of around 50 pm thus opening up the possibility of integration into lab-on-a-chip systems.

A double U-slot patch antenna with dual wideband characteristics

Microstrip patch antennas are strong candidates for use in many wireless communications applications. This paper proposes the use of a patch antenna with two U-shaped slots to achieve dual band operation. A thick substrate helps broaden the individual bandwidths. The antenna is designed based on extensive IE3D simulation studies. A prototype antenna is fabricated and experimentally verified for the required performance.

Dual Enlargement of Gold Nanoparticles: From Mechanism to Scanometric Detection of Pathogenic Bacteria

A mechanism of dual enlargement of gold nanoparticles (AuNPs) comprising two steps is described. In the first step, the AuNPs are enlarged by depositing Au atoms on their crystalline faces. In this process, the particles are not only enlarged but they are also observed to multiply: new Au nuclei are formed by the budding and division of the enlarged particles. In the second step, a silver enhancement is subsequently performed by the deposition of silver atoms on the enlarged and newly formed AuNPs to generate bimetallic Au@Ag core-shell structures. The dual nanocatalysis greatly enhances the electron density of the nanostructures, leading to a stronger intensity for colorimetric discrimination as well as better sensitivity for quantitative measurement. Based on this, a simple scanometric assay for the on-slide detection of the food-born pathogen Campylobacter jejuni is developed. After capturing the target bacteria, gold-tagged immunoprobes are added to create a signal on a solid substrate. The signal is then amplified by the dual enlargement process, resulting in a strong color intensity that can easily be recognized by the unaided eye, or measured by an inexpensive flatbed scanner. In this paper, dual nanocatalysis is reported for the first time. It provides a valuable mechanistic insight into the development of a simple and cost-effective detection format.

Dual-colour HER2/chromosome 17 chromogenic in situ hybridisation enables accurate assessment of HER2 genomic status in ovarian tumours

Ovarian cancer is a leading cause of gynaecological cancer-related morbidity and mortality. There has been increasing interest in the potential utility of anti-human epidermal growth factor receptor 2 (anti-HER2) agents in the treatment of this disease, with the attendant need to identify suitable predictive biomarkers of response to treatment.

Dual-colour HER2/chromosome 17 chromogenic in situ hybridisation assay enables accurate assessment of HER2 genomic status in gastric cancer and has potential utility in HER2 testing of biopsy samples

Determination of HER2 protein expression by immunohistochemistry (IHC) and genomic status by fluorescent in situ hybridisation (FISH) are important in identifying a subset of high HER2-expressing gastric cancers that might respond to trastuzumab. Although FISH is considered the standard for determination of HER2 genomic status, brightfield ISH is being increasingly recognised as a viable alternative. Also, the impact of HER2 protein expression/genomic heterogeneity on the accuracy of HER2 testing has not been well studied in the context of gastric biopsy samples.

Inference and Declaration of Independence in Task-Parallel Programs

The inherent difficulty of thread-based shared-memory programming has recently motivated research in high-level, task-parallel programming models. Recent advances of Task-Parallel models add implicit synchronization, where the system automatically detects and satisfies data dependencies among spawned tasks. However, dynamic dependence analysis incurs significant runtime overheads, because the runtime must track task resources and use this information to schedule tasks while avoiding conflicts and races.
We present SCOOP, a compiler that effectively integrates static and dynamic analysis in code generation. SCOOP combines context-sensitive points-to, control-flow, escape, and effect analyses to remove redundant dependence checks at runtime. Our static analysis can work in combination with existing dynamic analyses and task-parallel runtimes that use annotations to specify tasks and their memory footprints. We use our static dependence analysis to detect non-conflicting tasks and an existing dynamic analysis to handle the remaining dependencies. We evaluate the resulting hybrid dependence analysis on a set of task-parallel programs.