Generating Efficient Data Movement Code for Heterogeneous Architectures with Distributed-Memory

Programming for parallel architectures that do not have a shared address space is extremely difficult due to the need for explicit communication between memories of different compute devices. A heterogeneous system with CPUs and multiple GPUs, or a distributed-memory cluster are examples of such systems. Past works that try to automate data movement for distributed-memory architectures can lead to excessive redundant communication. In this paper, we propose an automatic data movement scheme that minimizes the volume of communication between compute devices in heterogeneous and distributed-memory systems. We show that by partitioning data dependences in a particular non-trivial way, one can generate data movement code that results in the minimum volume for a vast majority of cases. The techniques are applicable to any sequence of affine loop nests and works on top of any choice of loop transformations, parallelization, and computation placement. The data movement code generated minimizes the volume of communication for a particular configuration of these. We use a combination of powerful static analyses relying on the polyhedral compiler framework and lightweight runtime routines they generate, to build a source-to-source transformation tool that automatically generates communication code. We demonstrate that the tool is scalable and leads to substantial gains in efficiency. On a heterogeneous system, the communication volume is reduced by a factor of 11X to 83X over state-of-the-art, translating into a mean execution time speedup of 1.53X. On a distributed-memory cluster, our scheme reduces the communication volume by a factor of 1.4X to 63.5X over state-of-the-art, resulting in a mean speedup of 1.55X. In addition, our scheme yields a mean speedup of 2.19X over hand-optimized UPC codes.

Urea-Functionalized Self-Assembled Molecular Prism for Heterogeneous Catalysis in Water

Reaction of a ditopic urea ``strut'' (L-1) with cis-(tmen)Pd(NO3)(2) yielded a 3+3] self-assembled molecular triangle (T)L-1 = 1,4-di(4-pyridylureido)benzene; tmen = N,N,N',N'-tetrame-thylethane-1,2-diamine]. Replacing cis-(tmen)Pd(NO3)(2) in the above reaction with an equimolar mixture of Pd(NO3)(2) and a clip-type donor (L-2) yielded a template-free multicomponent 3D trigonal prism (P) decorated with multiple urea moieties L-2 = 3,3'-(1H-1,2,4-triazole-3,5-diyl)dipyridine]. This prism (P) was characterized by NMR. spectroscopy, and the structure was confirmed by X-ray crystallography. The P was employed as an effective hydrogen-bond-donor catalyst for Michael reactions of a series of water-insoluble nitro-olefins in an aqueous medium. The P showed better catalytic activity compared to the urea based ligand L-1 and the triangle T. Moreover, the confined nanospace of P in addition to large product outlet windows makes this 3D architecture a perfect molecular vessel to catalyze Diels-Alder reactions of 9-hydroxymethylanthracene with N-substituted maleimide in the aqueous medium. The present results demonstrate new observations on catalytic aqueous Diels-Alder and Michael reactions in heterogeneous fashion employing a discrete 3D architecture of Pd(II). The prism was recycled by simple filtration and reused several tithes without significant loss of activity.

Molecular Cage Impregnated Palladium Nanoparticles: Efficient, Additive-Free Heterogeneous Catalysts for Cyanation of Aryl Halides

Two shape-persistent covalent cages (CC1(r) and CC2(r)) have been devised from triphenyl amine-based trialdehydes and cyclohexane diamine building blocks utilizing the dynamic imine chemistry followed by imine bond reduction. The cage compounds have been characterized by several spectroscopic techniques which suggest that CC1(r) and CC2(r) are 2+3] and 8+12] self-assembled architectures, respectively. These state-of-the-art molecules have a porous interior and stable aromatic backbone with multiple palladium binding sites to engineer the controlled synthesis and stabilization of ultrafine palladium nanoparticles (PdNPs). As-synthesized cage-embedded PdNPs have been characterized by transmission electron microscopy (TEM), scanning electron microscopy (SEM), and powder X-ray diffraction (PXRD). Inductively coupled plasma optical emission spectrometry reveals that Pd@CC1(r) and Pd@CC2(r) have 40 and 25 wt% palladium loading, respectively. On the basis of TEM analysis, it has been estimated that as small as similar to 1.8 nm PdNPs could be stabilized inside the CC1(r), while larger CC2(r) could stabilize similar to 3.7 nm NPs. In contrast, reduction of palladium salts in the absence of the cages form structure less agglomerates. The well-dispersed cage-embedded NPs exhibit efficient catalytic performance in the cyanation of aryl halides under heterogeneous, additive-free condition. Moreover, these materials have excellent stability and recyclability without any agglomeration of PdNPs after several cycles.

Campaigning in Heterogeneous Social Networks: Optimal Control of SI Information Epidemics

We study the optimal control problem of maximizing the spread of an information epidemic on a social network. Information propagation is modeled as a susceptible-infected (SI) process, and the campaign budget is fixed. Direct recruitment and word-of-mouth incentives are the two strategies to accelerate information spreading (controls). We allow for multiple controls depending on the degree of the nodes/individuals. The solution optimally allocates the scarce resource over the campaign duration and the degree class groups. We study the impact of the degree distribution of the network on the controls and present results for Erdos-Renyi and scale-free networks. Results show that more resource is allocated to high-degree nodes in the case of scale-free networks, but medium-degree nodes in the case of Erdos-Renyi networks. We study the effects of various model parameters on the optimal strategy and quantify the improvement offered by the optimal strategy over the static and bang-bang control strategies. The effect of the time-varying spreading rate on the controls is explored as the interest level of the population in the subject of the campaign may change over time. We show the existence of a solution to the formulated optimal control problem, which has nonlinear isoperimetric constraints, using novel techniques that is general and can be used in other similar optimal control problems. This work may be of interest to political, social awareness, or crowdfunding campaigners and product marketing managers, and with some modifications may be used for mitigating biological epidemics.

The crystallization behavior of the metastable melts of poly(ethylene terephthalate) from the multiple melting process

During heating of semicrystalline PET, a metastable melt forms far below the equilibrium melting temperature. Crystallization kinetics of this metastable melt is discussed on the basis of DSC results. From the metastable melt almost one-dimensional growth of the crystal occurs through heterogeneous nucleation.

Multilevel mesh partitioning for heterogeneous communication networks

Multilevel algorithms are a successful class of optimisation techniques which address the mesh partitioning problem for distributing unstructured meshes onto parallel computers. They usually combine a graph contraction algorithm together with a local optimisation method which refines the partition at each graph level. To date these algorithms have been used almost exclusively to minimise the cut edge weight in the graph with the aim of minimising the parallel communication overhead, but recently there has been a perceived need to take into account the communications network of the parallel machine. For example the increasing use of SMP clusters (systems of multiprocessor compute nodes with very fast intra-node communications but relatively slow inter-node networks) suggest the use of hierarchical network models. Indeed this requirement is exacerbated in the early experiments with meta-computers (multiple supercomputers combined together, in extreme cases over inter-continental networks). In this paper therefore, we modify a multilevel algorithm in order to minimise a cost function based on a model of the communications network. Several network models and variants of the algorithm are tested and we establish that it is possible to successfully guide the optimisation to reflect the chosen architecture.

Assessing the potential of autonomous submarine gliders for ecosystem monitoring across multiple trophic levels (plankton to cetaceans) in shallow shelf seas

A combination of scientific, economic, technological and policy drivers is behind a recent upsurge in the use of marine autonomous systems (and accompanying miniaturized sensors) for environmental mapping and monitoring. Increased spatial–temporal resolution and coverage of data, at reduced cost, is particularly vital for effective spatial management of highly dynamic and heterogeneous shelf environments. This proof-of-concept study involves integration of a novel combination of sensors onto buoyancy-driven submarine gliders, in order to assess their suitability for ecosystem monitoring in shelf waters at a variety of trophic levels. Two shallow-water Slocum gliders were equipped with CTD and fluorometer to measure physical properties and chlorophyll, respectively. One glider was also equipped with a single-frequency echosounder to collect information on zooplankton and fish distribution. The other glider carried a Passive Acoustic Monitoring system to detect and record cetacean vocalizations, and a passive sampler to detect chemical contaminants in the water column. The two gliders were deployed together off southwest UK in autumn 2013, and targeted a known tidal-mixing front west of the Isles of Scilly. The gliders’ mission took about 40 days, with each glider travelling distances of >1000 km and undertaking >2500 dives to depths of up to 100 m. Controlling glider flight and alignment of the two glider trajectories proved to be particularly challenging due to strong tidal flows. However, the gliders continued to collect data in poor weather when an accompanying research vessel was unable to operate. In addition, all glider sensors generated useful data, with particularly interesting initial results relating to subsurface chlorophyll maxima and numerous fish/cetacean detections within the water column. The broader implications of this study for marine ecosystem monitoring with submarine gliders are discussed.

Assessing the potential of autonomous submarine gliders for ecosystem monitoring across multiple trophic levels (plankton to cetaceans) in shallow shelf seas

A combination of scientific, economic, technological and policy drivers is behind a recent upsurge in the use of marine autonomous systems (and accompanying miniaturized sensors) for environmental mapping and monitoring. Increased spatial–temporal resolution and coverage of data, at reduced cost, is particularly vital for effective spatial management of highly dynamic and heterogeneous shelf environments. This proof-of-concept study involves integration of a novel combination of sensors onto buoyancy-driven submarine gliders, in order to assess their suitability for ecosystem monitoring in shelf waters at a variety of trophic levels. Two shallow-water Slocum gliders were equipped with CTD and fluorometer to measure physical properties and chlorophyll, respectively. One glider was also equipped with a single-frequency echosounder to collect information on zooplankton and fish distribution. The other glider carried a Passive Acoustic Monitoring system to detect and record cetacean vocalizations, and a passive sampler to detect chemical contaminants in the water column. The two gliders were deployed together off southwest UK in autumn 2013, and targeted a known tidal-mixing front west of the Isles of Scilly. The gliders’ mission took about 40 days, with each glider travelling distances of >1000 km and undertaking >2500 dives to depths of up to 100 m. Controlling glider flight and alignment of the two glider trajectories proved to be particularly challenging due to strong tidal flows. However, the gliders continued to collect data in poor weather when an accompanying research vessel was unable to operate. In addition, all glider sensors generated useful data, with particularly interesting initial results relating to subsurface chlorophyll maxima and numerous fish/cetacean detections within the water column. The broader implications of this study for marine ecosystem monitoring with submarine gliders are discussed.

A Pipeline Interleaved Heterogeneous SIMD Soft Processor Array Architecture for MIMO-OFDM Detection

The most promising way to maintain reliable data transfer across the rapidly fluctuating channels used by next generation multiple-input multiple-output communications schemes is to exploit run-time variable modulation and antenna configurations. This demands that the baseband signal processing architectures employed in the communications terminals must provide low cost and high performance with runtime reconfigurability. We present a softcore-processor based solution to this issue, and show for the first time, that such programmable architectures can enable real-time data operation for cutting-edge standards
such as 802.11n; furthermore, by exploiting deep processing pipelines and interleaved task execution, the cost and performance of these architectures is shown to be on a par with traditional dedicated circuit based solutions. We believe this to be the first such programmable architecture to achieve this, and the combination of implementation efficiency and programmability makes this implementation style the most promising approach for hosting such dynamic architectures.

Accelerating Multiple Sequence Alignment with the Cell BE Processor

The Cell Broadband Engine (BE) Architecture is a new heterogeneous multi-core architecture targeted at compute-intensive workloads. The architecture of the Cell BE has several features that are unique in high-performance general-purpose processors, most notably the extensive support for vectorization, scratch pad memories and explicit programming of direct memory accesses (DMAs) and mailbox communication. While these features strongly increase programming complexity, it is generally claimed that significant speedups can be obtained by using Cell BE processors. This paper presents our experiences with using the Cell BE architecture to accelerate Clustal W, a bio-informatics program for multiple sequence alignment. We report on how we apply the unique features of the Cell BE to Clustal W and how important each is in obtaining high performance. By making extensive use of vectorization and by parallelizing the application across all cores, we demonstrate a speedup of 24.4 times when using 16 synergistic processor units on a QS21 Cell Blade compared to single-thread execution on the power processing unit. As the Cell BE exploits a large number of slim cores, our highly optimized implementation is just 3.8 times faster than a 3-thread version running on an Intel Core2 Duo, as the latter processor exploits a small number of fat cores.

The ParaPhrase Project:Parallel patterns for adaptive heterogeneous multicore systems

This paper describes the ParaPhrase project, a new 3-year targeted research project funded under EU Framework 7 Objective 3.4 (Computer Systems), starting in October 2011. ParaPhrase aims to follow a new approach to introducing parallelism using advanced refactoring techniques coupled with high-level parallel design patterns. The refactoring approach will use these design patterns to restructure programs defined as networks of software components into other forms that are more suited to parallel execution. The programmer will be aided by high-level cost information that will be integrated into the refactoring tools. The implementation of these patterns will then use a well-understood algorithmic skeleton approach to achieve good parallelism. A key ParaPhrase design goal is that parallel components are intended to match heterogeneous architectures, defined in terms of CPU/GPU combinations, for example. In order to achieve this, the ParaPhrase approach will map components at link time to the available hardware, and will then re-map them during program execution, taking account of multiple applications, changes in hardware resource availability, the desire to reduce communication costs etc. In this way, we aim to develop a new approach to programming that will be able to produce software that can adapt to dynamic changes in the system environment. Moreover, by using a strong component basis for parallelism, we can achieve potentially significant gains in terms of reducing sharing at a high level of abstraction, and so in reducing or even eliminating the costs that are usually associated with cache management, locking, and synchronisation. © 2013 Springer-Verlag Berlin Heidelberg.

Cooperative positioning for heterogeneous wireless systems

Future emerging market trends head towards positioning based services placing a new perspective on the way we obtain and exploit positioning information. On one hand, innovations in information technology and wireless communication systems enabled the development of numerous location based applications such as vehicle navigation and tracking, sensor networks applications, home automation, asset management, security and context aware location services. On the other hand, wireless networks themselves may bene t from localization information to improve the performances of di erent network layers. Location based routing, synchronization, interference cancellation are prime examples of applications where location information can be useful. Typical positioning solutions rely on measurements and exploitation of distance dependent signal metrics, such as the received signal strength, time of arrival or angle of arrival. They are cheaper and easier to implement than the dedicated positioning systems based on ngerprinting, but at the cost of accuracy. Therefore intelligent localization algorithms and signal processing techniques have to be applied to mitigate the lack of accuracy in distance estimates. Cooperation between nodes is used in cases where conventional positioning techniques do not perform well due to lack of existing infrastructure, or obstructed indoor environment. The objective is to concentrate on hybrid architecture where some nodes have points of attachment to an infrastructure, and simultaneously are interconnected via short-range ad hoc links. The availability of more capable handsets enables more innovative scenarios that take advantage of multiple radio access networks as well as peer-to-peer links for positioning. Link selection is used to optimize the tradeo between the power consumption of participating nodes and the quality of target localization. The Geometric Dilution of Precision and the Cramer-Rao Lower Bound can be used as criteria for choosing the appropriate set of anchor nodes and corresponding measurements before attempting location estimation itself. This work analyzes the existing solutions for node selection in order to improve localization performance, and proposes a novel method based on utility functions. The proposed method is then extended to mobile and heterogeneous environments. Simulations have been carried out, as well as evaluation with real measurement data. In addition, some speci c cases have been considered, such as localization in ill-conditioned scenarios and the use of negative information. The proposed approaches have shown to enhance estimation accuracy, whilst signi cantly reducing complexity, power consumption and signalling overhead.

Real-time scheduling with resource sharing on heterogeneous multiprocessors

Consider the problem of scheduling a task set τ of implicit-deadline sporadic tasks to meet all deadlines on a t-type heterogeneous multiprocessor platform where tasks may access multiple shared resources. The multiprocessor platform has m k processors of type-k, where k∈{1,2,…,t}. The execution time of a task depends on the type of processor on which it executes. The set of shared resources is denoted by R. For each task τ i , there is a resource set R i ⊆R such that for each job of τ i , during one phase of its execution, the job requests to hold the resource set R i exclusively with the interpretation that (i) the job makes a single request to hold all the resources in the resource set R i and (ii) at all times, when a job of τ i holds R i , no other job holds any resource in R i . Each job of task τ i may request the resource set R i at most once during its execution. A job is allowed to migrate when it requests a resource set and when it releases the resource set but a job is not allowed to migrate at other times. Our goal is to design a scheduling algorithm for this problem and prove its performance. We propose an algorithm, LP-EE-vpr, which offers the guarantee that if an implicit-deadline sporadic task set is schedulable on a t-type heterogeneous multiprocessor platform by an optimal scheduling algorithm that allows a job to migrate only when it requests or releases a resource set, then our algorithm also meets the deadlines with the same restriction on job migration, if given processors 4×(1+MAXP×⌈|P|×MAXPmin{m1,m2,…,mt}⌉) times as fast. (Here MAXP and |P| are computed based on the resource sets that tasks request.) For the special case that each task requests at most one resource, the bound of LP-EE-vpr collapses to 4×(1+⌈|R|min{m1,m2,…,mt}⌉). To the best of our knowledge, LP-EE-vpr is the first algorithm with proven performance guarantee for real-time scheduling of sporadic tasks with resource sharing on t-type heterogeneous multiprocessors.

Adaptation of an commercially available stabilised R/C helicopter to a fully autonomous surveillance UAV

This paper presents the development of an autonomous surveillance UAV that competed in the Ministry of Defence Grand Challenge 2008. In order to focus on higher-level mission control, the UAV is built upon an existing commercially available stabilised R/C helicopter platform. The hardware architecture is developed to allow for non-invasion integration with the existing stabilised platform, and to enable to the distributed processing of closed loop control and mission goals. The resulting control system proved highly successful and was capable of flying within 40knott gusts. The software and safety architectures were key to the success of the research and also hold the potential for use in the development of more complex system comprising of multiple UAVs.

Modelling of urban sensible heat flux at multiple spatial scales : a demonstration using airborne hyperspectral imagery of Shanghai and a temperature–emissivity separation approach

Urbanization related alterations to the surface energy balance impact urban warming (‘heat islands’), the growth of the boundary layer, and many other biophysical processes. Traditionally, in situ heat flux measures have been used to quantify such processes, but these typically represent only a small local-scale area within the heterogeneous urban environment. For this reason, remote sensing approaches are very attractive for elucidating more spatially representative information. Here we use hyperspectral imagery from a new airborne sensor, the Operative Modular Imaging Spectrometer (OMIS), along with a survey map and meteorological data, to derive the land cover information and surface parameters required to map spatial variations in turbulent sensible heat flux (QH). The results from two spatially-explicit flux retrieval methods which use contrasting approaches and, to a large degree, different input data are compared for a central urban area of Shanghai, China: (1) the Local-scale Urban Meteorological Parameterization Scheme (LUMPS) and (2) an Aerodynamic Resistance Method (ARM). Sensible heat fluxes are determined at the full 6 m spatial resolution of the OMIS sensor, and at lower resolutions via pixel aggregation and spatial averaging. At the 6 m spatial resolution, the sensible heat flux of rooftop dominated pixels exceeds that of roads, water and vegetated areas, with values peaking at ∼ 350 W m− 2, whilst the storage heat flux is greatest for road dominated pixels (peaking at around 420 W m− 2). We investigate the use of both OMIS-derived land surface temperatures made using a Temperature–Emissivity Separation (TES) approach, and land surface temperatures estimated from air temperature measures. Sensible heat flux differences from the two approaches over the entire 2 × 2 km study area are less than 30 W m− 2, suggesting that methods employing either strategy maybe practica1 when operated using low spatial resolution (e.g. 1 km) data. Due to the differing methodologies, direct comparisons between results obtained with the LUMPS and ARM methods are most sensibly made at reduced spatial scales. At 30 m spatial resolution, both approaches produce similar results, with the smallest difference being less than 15 W m− 2 in mean QH averaged over the entire study area. This is encouraging given the differing architecture and data requirements of the LUMPS and ARM methods. Furthermore, in terms of mean study QH, the results obtained by averaging the original 6 m spatial resolution LUMPS-derived QH values to 30 and 90 m spatial resolution are within ∼ 5 W m− 2 of those derived from averaging the original surface parameter maps prior to input into LUMPS, suggesting that that use of much lower spatial resolution spaceborne imagery data, for example from Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) is likely to be a practical solution for heat flux determination in urban areas.