Yuxin Zheng

Partitioned Global Address Space languages are convenient for expressing algorithms with large, random-access data, and they have proven to provide high performance and scalability through lightweight one-sided communication and locality control. While very convenient for moving data around the system, PGAS languages have taken different views on the model of computation, with the static Single Program Multiple Data model providing the best scalability. In this paper we present UPC++, a PGAS extension for C++ that has three main objectives: 1) to provide an object-oriented PGAS programming model in the context of the popular C++ language, 2) to add useful parallel programming idioms unavailable in UPC, such as asynchronous remote function invocation and multidimensional arrays, to support complex scientific applications, 3) to offer an easy on-ramp to PGAS programming through interoperability with other existing parallel programming systems. We implement UPC++ with a 'compiler-free' approach using C++ templates and runtime libraries. We borrow heavily from previous PGAS languages and describe the design decisions that led to this particular set of language features, providing significantly more expressiveness than UPC with very similar performance characteristics. We evaluate the programmability and performance of UPC++ using five benchmarks on two representative supercomputers, demonstrating that UPC++ can deliver excellent performance at large scale up to 32K cores while offering PGAS productivity features to C++ applications. © 2014 IEEE.

Copyright 2014 ACM 978-1-4503-3247-7/14/10$15.00.Partitioned Global Address Space languages and one-sided communication enable application developers to select the communication paradigm that balances the performance needs of applications with the productivity desires of programmers. In this paper, we evaluate three different one-sided communication paradigms in the context of geometric multigrid using the miniGMG benchmark. Although miniGMG's static, regular,and predictable communication does not exploit the ultimate potential of PGAS models, multigrid solvers appear in many contemporary applications and represent one of the most important communication patterns. We use UPC++, a PGAS extension of C++, as the vehicle for our evaluation, though our work is applicable to any of the existing PGAS languages and models. We compare performance with the highly tuned MPI baseline, and the results indicate that the most promising approach towards achieving performance and ease of programming is to use high-level abstractions, such as the multidimensional arrays provided by UPC++, that hide data aggregation and messaging in the runtime library. 2014 Association for Computing Machinery.

The human cationic antimicrobial protein LL-37 is a multifunctional host defense peptide with a wide range of immunomodulatory activities. Previous work has shown that LL-37 exerts both pro- and anti-inflammatory effects. The role of mitochondria in the skin inflammatory effects of LL-37 has not been well studied. Therefore, our aim was to investigate the immunomodulatory effect of LL-37 in HaCaT cells and to delineate the underlying mechanisms related to mitochondrial function. Immunohistochemistry results from tissue microarrays showed strong cytoplasmic LL-37 staining in inflammatory cells in chronic dermatic inflammation. Using exogenous LL-37 stimulation and LL-37 knockdown and overexpression, LL-37 was demonstrated to dramatically reduce the mRNA levels and protein secretion of inflammatory cytokines including IL-6, IL-8, IL-1α and tumor necrosis factor-α, which are induced by lipopolysaccharides. The anti-inflammatory effects of LL-37 are dependent upon its ability to increase mitochondrial biogenesis and to maintain mitochondrial homeostasis. Furthermore, we observed that LL-37 enhances the LPS-induced phosphorylation of extracellular signal-regulated kinase and mammalian target of rapamycin. The mTOR inhibitor rapamycin can neutralize the protective effects of LL-37 on mitochondria. In conclusion, these results suggest that high LL-37 expression levels correlate with chronic skin inflammation; mitochondrial dysfunction occurs in HaCaT cells during inflammation; and LL-37 attenuates inflammatory impairment by stimulating mitochondrial biogenesis and protecting mitochondrial function, which are dependent upon mTOR signaling. These findings provide new insights into targeting mitochondria with LL-37 to prevent skin inflammatory reactions. © 2014 Elsevier Ltd.

Both SKP2 and transforming growth factor-β1 play important roles in cancer metastasis through different mechanisms: TGF-β1 via induction of epithelial-mesenchymal transition and SKP2 via downregulating p27kip1. Recent studies indicated that c-Myc and Akt1 were active players in metastasis. In this study we demonstrated a crosstalk between these pathways. Specifically, we found that TGF-β1 treatment increased SKP2 expression accompanied with increased phosphorylation of Akt1 and c-Myc protein accumulation during EMT. We demonstrated that Akt1 was required for TGF-β1-mediated SKP2 upregulation and that c-Myc transcription factor specifically bound to the promoter of SKP2 for its enhanced transcription. Analysis of 25 samples of normal human skin, nevi, and melanomas revealed a positive correlation between c-Myc and SKP2 accumulation. Furthermore, accumulation of SKP2 and c-Myc proteins was significantly higher in metastatic melanoma samples as compared with that in primary melanomas, which again was higher than that in normal skin or nevi. In summary, our results integrated TGF-β1 signals to SKP2 via Akt1 and c-Myc during EMT, and provided, to our knowledge, a previously unreported mechanistic molecular event for TGF-β1-induced metastasis in human melanoma. © 2014 The Society for Investigative Dermatology.

Copyright © 2014 ACM 978-1-4503-2937-8/14/06. 15.00.Multidimensional arrays are an important data structure in many scientific applications. Unfortunately, built-in support for such arrays is inadequate in C++, particularly in the distributed setting where bulk communication operations are required for good performance. In this paper, we present a multidimensional library for partitioned global address space programs, supporting the one-sided remote access and bulk operations of the PGAS model. The library is based on Titanium arrays, which have proven to provide good productivity and performance. These arrays provide a local view of data, where each rank constructs its own portion of a global data structure, matching the local view of execution common to PGAS programs and providing maximum flexibility in structuring global data. Unlike Titanium, which has its own compiler with array-specific analyses, optimizations, and code generation, we implement multidimensional arrays solely through a C++ library. The main goal of this effort is to provide a library-based implementation that can match the productivity and performance of a compiler-based approach. We implement the array library as an extension to UPC++, a C++ library for PGAS programs, and we extend Titanium arrays with specializations to improve performance. We evaluate the array library by porting four Titanium benchmarks to UPC++, demonstrating that it can achieve up to 25% better performance than Titanium without a significant increase in programmer effort..

Partitioned Global Address Space languages offer programmers the convenience of a shared memory programming style combined with locality control necessary to run on large-scale distributed memory systems. Even within a PGAS language programmers often need to perform global communication operations such as broadcasts or reductions, which are best performed as collective operations in which a group of threads work together to perform the operation. In this paper we consider the problem of implementing collective communication within PGAS languages and explore some of the design trade-offs in both the interface and implementation. In particular, PGAS collectives have semantic issues that are different than in send-receive style message passing programs, and different implementation approaches that take advantage of the one-sided communication style in these languages. We present an implementation framework for PGAS collectives as part of the GASNet communication layer, which supports shared memory, distributed memory and hybrids. The framework supports a broad set of algorithms for each collective, over which the implementation may be automatically tuned. Finally, we demonstrate the benefit of optimized GASNet collectives using application benchmarks written in UPC, and demonstrate that the GASNet collectives can deliver scalable performance on a variety of state-of-the-art parallel machines including a Cray XT4, an IBM BlueGene/P, and a Sun Constellation system with InfiniBand interconnect. © 2011 Elsevier B.V. All rights reserved.

Dynamic interaction is said to occur when two significanrly different fields A and B come into relation, and their interaction is dynamic in the sense that at first the flow of ideas is principally from A to B, but later ideas from B come to influence A. Two examples are given of dynamic interactions with the philosophy of mathematics. The first is with philosophy of scicnce, and thc sccond with computer science. Theanalysis cnables Lakatos to be charactcrised as thc first to devclop the philosophy of mathematics using ideas taken from thc philosophy of science.

BackgroundExercise is one of the effective ways to improve cognition. Different forms of exercises, such as aerobic exercise, resistance exercise, and coordination exercise, have different effects on the improvement of cognitive impairment. In recent years, exergames based on Non-Immersive Virtual Reality have been widely used in entertainment and have gradually been applied to clinical rehabilitation. However, the mechanism of NIVR-Exergames on improving motor cognition has not been clarified. Therefore, the aim of this study is to find whether NIVR-Exergames result in a better neural response mechanism to improve the area of the cerebral cortex related to motor cognition under functional near-infrared spectroscopy dynamic monitoring in comparison with resistance exercise.MethodsA cross-over study design was adopted in this study, and 15 healthy young subjects were randomly divided into group A and group B according to a computerized digital table method. Task 1 was an NIVR-Exergame task, and Task 2 was resistance band stretching. Group A first performed Task 1, rested for 30 min, and then performed Task 2. Group B had the reverse order. The fNIRS test was synchronized in real time during exercise tasks, and heart rate measurements, blood pressure measurements, and 2-back task synchronization fNIRS tests were performed at baseline, Post-task 1, and Post-task 2. The primary outcomes were beta values from the general linear model in different regions of interest, and the secondary outcomes were heart rate, blood pressure, reaction time of 2-back, and accuracy rate of 2-back.ResultsThe activation differences of Task 1 and Task 2 in the right premotor cortex and the left PMC were statistically significant. There were statistically significant differences in the activation of the right supplementary motor area, left dorsolateral prefrontal cortex, left and right PMC between baseline and Post-task 1. The differences in systolic pressure between the two groups at three time points among women were statistically significant.ConclusionIn this study, we found that NIVR-Exergames combined with motor and challenging cognitive tasks can promote the activation of SMA, PMC and DLPFC in healthy young people compared with resistance exercise alone, providing compelling preliminary evidence of the power for the rehabilitation of motor and cognitive function in patients with central nervous system diseases.