High-level data-structures are an important foundation for most applications. With the rise of multicores, there is a trend of supporting data-parallel collection operations in general purpose programming languages. However, these operations often incur high-level abstraction and scheduling penalties. We present a generic data-parallel collections design based on work-stealing for shared-memory architectures that overcomes abstraction penalties through call site specialization of data-parallel operation instances.
Moreover, we introduce work-stealing iterators that allow more fine-grained and efficient work-stealing. By eliminating abstraction penalties and making work-stealing data-structure-aware we achieve several dozen times better performance compared to existing JVM-based approaches.
Continue Reading
This paper is to review the research activities at City University London in the development of a suite of optical fibre-based chemical sensors, including pH, humidity and heavy metal sensors,…
Ultra-wide band (UWB) pulse Doppler radars provide range-time-frequency information which enables the target localization and vital sign monitoring of a subject. One challenge for UWB radar systems is accurately detecting…
A reconfigurable microwave photonic filter based on a Brillouin-assisted narrowband rectangular optical amplifier with a constant gain and a linear phase response is proposed. The optical amplifier is achieved by…
The second order transverse autocorrelation technique, using a nonlinear crystal with a random-size and distribution of antiparallel nonlinear domains, has been recently proved to be an effective method for ultrashort…
In this paper, we focus on subcarrier and power allocation for an orthogonal frequency division multiple access (OFDMA) full-duplex (FD) system. A three-step algorithm is proposed to maximize the sum-rate…
A simple continuous fabrication process for graded-index polymer optical fibers is presented that does not rely on dopants to form the refractive-index profile. A conventional melt-spinning process is followed by…
The increasing volumes of relational data let us find an alternative to cope with them. The Hadoop framework - which is an open source project based on the MapReduce paradigm…
In this paper, the outage probability and average bit error rate of a multiuser, multiple input multiple output (MIMO) systems has been incorporated with transmit antenna selection maximal ratio combining…
The optical (such as absorption coefficient, linear refractive index, real and imaginary parts of dielectric constants and energy band gap) and thermal (such as diffusivities, effectivities and conductivities) properties of…