Compiler and runtime analysis for efficient communication in data intensive applications

Abstract: Processing and analyzing large volumes of data plays an increasingly important role in many domains of scientific research. We are developing a compiler that processes data intensive applications written in a dialect of Java and compiles them for efficient execution on distributed memory parallel machines. In this paper, we focus on the problem of generating correct and efficient communication for data intensive applications. We present static analysis techniques for 1) extracting a global reduction function from a data parallel loop, and 2) determining if a subscript function is monotonic. We also present a runtime technique for reducing the volume of communication during the global reduction phase. We have experimented with two data intensive applications to evaluate the efficacy of our techniques. Our results show that 1) our techniques for extracting global reduction functions and establishing monotonicity of subscript functions can successfully handle these applications, 2) significant reduction in communication volume and execution times is achieved through our runtime analysis technique, 3) runtime communication analysis is critical for achieving speedups on parallel configurations.

Current microprocessors aggressively exploit instruction-level parallelism (ILP) through techniques such as multiple issue, dynamic scheduling, and non-blocking reads. Recent work has shown that memory latency remains a significant performance bottleneck for shared-memory multiprocessor systems built of such processors.This paper provides the first study of the effectiveness of software-controlled non-binding prefetching in shared memory multiprocessors built of state-of-the-art ILP-based processors. We find that software prefetching results in significant reductions in execution time (12% to 31%) for three out of five applications on an ILP system. However, compared to previous-generation system, software prefetching is significantly less effective in reducing the memory stall component of execution time on an ILP system. Consequently, even after adding software prefetching, memory stall time accounts for over 30% of the total execution time in four out of five applications on our ILP system.This paper also investigates the interaction of software prefetching with memory consistency models on ILP-based multiprocessors. In particular, we seek to determine whether software prefetching can equalize the performance of sequential consistency (SC) and release consistency (RC). We find that even with software prefetching, for three out of five applications, RC provides a significant reduction in execution time (15% to 40%) compared to SC.

The recent trend towards tile-based manycore architectures has helped to tackle the memory wall by physically distributing memories and processing nodes. However, this introduced a data-to-task locality challenge and inter-tile communication thus often imposes significant software overhead. Thus, we proposed software-defined hardware-managed SHARQ queues that enable efficient inter-tile communication by leveraging user-defined queues with arbitrarily sized elements. To ensure (remote) processing of queued elements, SHARQ introduces an optional handler task, which is scheduled by hardware on demand. Queue management, intra- and inter-tile data transfer, and handler task invocation are entirely handled by hardware. Only rare tasks, like the dynamic queue creation at run-time, are performed in software. DySHARQ, an extension of SHARQ, enables dynamic and concurrent queue memory management and queue length adjustments to be able to adapt to application and resource requirement changes. The DySHARQ hardware is able to monitor the queue memory requirements at run-time and conditionally schedules a software-defined memory management task. It further optimizes the hardware-software interaction for local queue operations. We integrated DySHARQ into the MPI library used by the NAS benchmarks. The evaluation shows a reduction in execution time by up to 43% (compared to software) for the communication intense IS kernel in a 4 $$\times$$ 4 tile design on an FPGA platform with a total of 80 LEON3 cores. The dynamic memory management reduces the memory footprint by 3.75 $$\times$$ in a 2 $$\times$$ 2 design.

The past decade has witnessed an explosive increase in the scale, intensity and sophistication of cyber attacks. While software vendors have significantly increased their efforts on security, they are almost always playing catch up. As a result, security-conscious organizations and individuals have come to expect their system administrators to deploy an array of tools and techniques to stay a step ahead of the hackers. While developer-oriented security tools rely mainly on static analysis, runtime analysis and policy enforcement are the mechanisms of choice in administrator-oriented tools. Runtime techniques offer increased automation and precision over static analysis, thus addressing the needs of administrators who don't have the time or resources needed to acquire extensive knowledge about the internals of a software system. In this talk, I will begin by summarizing some of the significant advances that have been achieved in the past few years in the context of software vulnerability mitigation, including buffer overflow defenses, and more recently, the impressive results that have been achieved using dynamic information-flow analysis for blocking the most popular exploits today, including SQL and command injection and cross-site scripting. I will then proceed to describe dynamic analysis and enforcement techniques aimed at another high-profile security problem faced today, namely, malware defense. Our initial target in this regard has been on dynamic analysis techniques for extracting high-level models of program behavior. These models could be used in a variety of applications such as intrusion detection, vulnerability analysis and security policy verification. More recently, interesting advances have been made in the context of security policy development, where a combination of static and dynamic analysis techniques have been developed to synthesize low-level, enforceable policies that achieve a high-level goal such as protecting system integrity. Finally, I will conclude the talk with a discussion of some research opportunities and challenges in software security.

The protein threading problem is the problem of determining the three-dimensional structure of a given but arbitrary protein sequence from a set of known structures of other proteins. This problem is known to be NP-hard and current computational approaches to threading are time-consuming and data-intensive. In this paper, we propose an evolution strategy for protein threading. We also developed two parallel approaches for fast threading. We have obtained at least better results than current comparable approaches, as well as significant reduction in execution time.

In this work, we performed a numerical simulation of the non-isothermal flow in an oil reservoir. Also, we used the OpenACC API to parallelize specific parts of the original code, which allowed the simultaneous execution of different tasks on a NVIDIA GTX 970 G1 video card in a shared memory architecture. The problem was studied using a vertical producer well and static heating wells in a two-dimensional domain. We used the Control Volume Finite Difference (CVFD) method to discretize the governing equations and the Conjugate Gradients method to obtain the solutions (pressure and temperature) of the systems of algebraic equations. As a result, concerning computational performance, a significant reduction in execution time was obtained with the parallelized version.

The recent data deluge needing to be processed represents one of the major challenges in computational field. Available high-performance computing (HPC) systems can be very useful for solving this problem when data can be divided in chunks that can be processed in parallel. However, due to intrinsic characteristics of data-intensive problems, these applications can present huge load imbalances, and it can be difficult to efficiently use the available resources. This work proposes a strategy for dynamically analyzing and tuning the partition factor used to generate the data chunks. With the aim of decreasing the load imbalance and therefore the overall execution time, this strategy divides the data chunks with the biggest computation times and gathers contiguous chunks with the smallest computation times. The criteria to divide or join chunks are based on the chunks' associated execution time (average and standard deviation) and the number of processing nodes being used. We have evaluated our strategy by using simulation, and a real data-intensive application. Applying our strategy, we have obtained promising results since we have improved up to 55% the total execution time.

Heterogeneous manycore architectures are the key to efficiently execute compute- and data-intensive applications. Through-silicon-via (TSV)-based 3D manycore system is a promising solution in this direction as it enables the integration of disparate computing cores on a single system. Recent industry trends show the viability of 3D integration in real products (e.g., Intel Lakefield SoC Architecture, the AMD Radeon R9 Fury X graphics card, and Xilinx Virtex-7 2000T/H580T, etc.). However, the achievable performance of conventional TSV-based 3D systems is ultimately bottlenecked by the horizontal wires (wires in each planar die). Moreover, current TSV 3D architectures suffer from thermal limitations. Hence, TSV-based architectures do not realize the full potential of 3D integration. Monolithic 3D (M3D) integration, a breakthrough technology to achieve “More Moore and More Than Moore,” opens up the possibility of designing cores and associated network routers using multiple layers by utilizing monolithic inter-tier vias (MIVs) and hence, reducing the effective wire length. Compared to TSV-based 3D integrated circuits (ICs), M3D offers the “true” benefits of vertical dimension for system integration: the size of an MIV used in M3D is over 100 × smaller than a TSV. This dramatic reduction in via size and the resulting increase in density opens up numerous opportunities for design optimizations in 3D manycore systems: designers can use up to millions of MIVs for ultra-fine-grained 3D optimization, where individual cores and routers can be spread across multiple tiers for extreme power and performance optimization. In this work, we demonstrate how M3D-enabled vertical core and uncore elements offer significant performance and thermal improvements in manycore heterogeneous architectures compared to its TSV-based counterpart. To overcome the difficult optimization challenges due to the large design space and complex interactions among the heterogeneous components (CPU, GPU, Last Level Cache, etc.) in a M3D-based manycore chip, we leverage novel design-space exploration algorithms to trade off different objectives. The proposed M3D-enabled heterogeneous architecture, called HeM3D , outperforms its state-of-the-art TSV-equivalent counterpart by up to 18.3% in execution time while being up to 19°C cooler.

Data-intensive applications process large volumes of data using a parallel processing method. MapReduce is a programming model designed for data-intensive applications for massive data sets and an execution framework for large-scale data processing on clusters of commodity servers. While fault tolerance, easy programming structure, and high scalability are considered strong points of MapReduce; however its configuration parameters must be fine-tuned to the specific deployment, which makes it more complex in configuration and performance. This paper explains tuning of the Hadoop configuration parameters, which directly affect MapReduce's job workflow performance under various conditions to achieve maximum performance. On the basis of the empirical data we collected, it became apparent that three main methodologies can affect the execution time of MapReduce running on cluster systems. Therefore, in this paper, we present a model that consists of three main modules: (1) Extending a data redistribution technique in order to find the high-performance nodes, (2) Utilizing the number of map/reduce slots in order to make it more efficient in terms of execution time, and (3) Developing a new hybrid routing schedule shuffle phase in order to define the scheduler task while memory management level is reduced. © 2013 Wiley Periodicals, Inc.

Background: Stroke is a leading cause of major long-term disability worldwide. Despite availability of IV tPA and mechanical thrombectomy, over half of treated patients have significant disability. Cell based therapy is a promising in pre-clinical studies in improving stroke recovery. Brain infarct volume is a major predictor of stroke outcome and may decrease with this treatment. We hypothesized that patients treated with intra-arterial stem cell therapy will have a significant reduction in stroke volume compared to non-treated patients. Methods: The RECOVER-Stroke trial is a randomized, sham-controlled study to determine the safety and efficacy of processed autologous bone marrow cells (ALD-401) injection via intracarotid infusion in ischemic stroke patients. Ischemic stroke patients were double-blind randomized into 2 groups, receiving ALD-401 injection or sham procedure between 9-19 days post anterior circulation ischemic stroke. All enrolled patients at 1 of 8 sites in the trial were included in this analysis. Two blinded physicians used OSIRIX software to analyze stroke volume on DWI/ 24h CT brain and MRI at 90, 180, 365 days post intervention. Volumes were compared between the 2 groups at these timepoints. Results: A total of 11 patients at our center (out of 48 multicenter patients) were included in this study, 6 patients receiving sham procedure and 5 patients receiving ALD-401. The two groups did not show any significant difference in their baseline variables including age, Mean baseline stroke volumes were not significantly different between ALD-401 (103.9±96.2ml) and sham groups (70.2±46ml), p=0.53. The stroke volume reduction in the ALD-401 was 22.9±15.5ml vs. 4.04±12.5ml in the sham group (p=0.046). On day 180 and 365, both groups did not show any further significant stroke volume reduction from day 90. Conclusion: In this exploratory analysis from one center in the multi-center RECOVER-Stroke trial, intra-arterial stem cell injection with ALD-401 was associated with significant reduction in stroke volume at 90 days post therapy compared to sham procedure in subacute ischemic stroke patients.

A parallel rough set based dependency calculation method for efficient feature selection

Efficient implementation of DSP applications is critical for embedded systems. However, current applications written in C, make extensive use of pointer arithmetic making compiler analysis and optimisation difficult. This paper presents a method for conversion of a restricted class of pointer-based memory accesses typically found in DSP codes into array accesses with explicit index functions. C programs with pointer accesses to array elements, data independent pointer arithmetic and structured loops can be converted into semantically equivalent representations with explicit array accesses. This technique has been applied to several DSPstone benchmarks on three different processors where initial results show that this technique can give on average a 11.95 % reduction in execution time after transforming pointer-based array accesses into explicit array accesses.

World Health Organization (W.H.O) has coined the word “Infodemic” to refer to the dissemination of fake news during this pandemic, which is considered to be as harmful as the virus itself. Verifying the information available on the internet is a prerequisite to ensuring the ecosystem is maintained which is the driving force behind this work. The primary goal of this study is to address the problem of time-consuming automatically voluminous fake news detection of certain data and consider the uncertainty of data from causal relations using a rich feature set. This research produces significant feature reduction for reduced time and improved accuracy by filtering out significant features using recursive feature selection (RFE). The retained features by the RFE algorithm are also compared to a standard statistical measure of Pearson’s correlation to ensure no information loss while reducing features. The suggested methodology has also defined appropriate class output assurance levels and accurate prediction ambiguity for the fake identification jobs. Comparative analysis for existing methods for feature selection is performed. The result of experimentation testifies the improvement of a 6% increase in precision and a 97% reduction in execution time.

In this paper we consider the problem of searching for an arbitrarily smart and fast evader in a large environment with a team of unmanned aerial vehicles (UAVs) while providing guarantees of detection. Our emphasis is on the fast execution of efficient search strategies that minimize the number of UAVs and the search time. We present the first approach for computing fast guaranteed search strategies utilizing additional searchers to speed up the execution time and thereby enabling large scale UAV search. In order to scale to very large environments when using UAVs one would either have to overcome the energy limitations of UAVs or pay the cost of utilizing additional UAVs to speed up the search. Our approach is based on coordinating UAVs on sweep lines, covered by the UAV sensors, that move simultaneously through an environment. We present some simulation results that show a significant reduction in execution time when using multiple UAVs and a demonstration of a real system with three AR.Drone 2.0.

Glioblastoma multiforme (GBM) remains a challenging malignancy due to intrinsic resistance to therapy. Pre-clinical mouse models which recapitulate the human disease provide an opportunity to identify and test novel agents that could be used alone or in combination with other systemic agents and/or radiation therapy. Here we describe the use of a genetically engineered murine GBM model [Ink4a-Arf-/- PtenloxP/loxp/ Ntv-a RCAS/PDGF(+)/Cre(+)] and functional magnetic resonance imaging (MRI) to evaluate systemic agents alone or in combination with radiation. The current standard therapy for patients bearing GBMs is Temozolomide (TMZ) combined with radiation yet recent analysis of O-6-methylguanine-DNA methyltransferase (MGMT) status in GBMs has linked unmethylated MGMT promoter with poor outcome. Thus we evaluated Gemcitabine (GEM) as a novel therapy for GBM concurrent with radiation and compared it to standard therapy. GEM, a deoxycytidine analog, is a potent radiosensitizer that demonstrates cytotoxic, anti-clonogenic, and radiosensitizing effects in GBM cell lines. Therapeutic efficacy was quantified by contrast-enhanced MRI and diffusion-weighted MRI (DW-MRI) for growth rate and tumor cellularity, respectively. Mice treated with GEM, TMZ or radiation alone showed a significant reduction in tumor volume compared to control mice as early as three days post treatment start. Both combined treatment (GEM+IR and TMZ+IR) resulted in improved effectiveness over single therapies. A significant reduction in tumor volume compared to GEM, TMZ or IR groups was observed. Furthermore, both combination therapies significantly prolonged survival compared to the other groups as early as day 7 after start of treatment. There was no significant difference between both combined treatments either in tumor growth or in survival. ADC values were found to increase immediately upon administration of GEM in both alone and combination therapies. In contrast, a gradual increase in ADC values was observed in tumors treated with IR and TMZ, both alone and in combination.GEM concurrent with radiation resulted in significant reduction in tumor volume and prolonged survival in this clinically-relevant GBM mouse model. GEM combined with radiation had the same effect GBM as compared to TMZ combined with radiation. We also demonstrated the successful implementation of an imaging biomarker surrogate (ADC) which correlated with efficacy. Ink4a-Arf-/- PtenloxP/loxP/ Ntv-a RCAS/PDGF(+)/Cre(+) mice serve as a valuable pre-clinical GBM model with the potential to identify novel therapeutics for early phase I/II testing for GBM patients. Based on these results, GEM combined with IR appears to be a suitable alternative to treat tumors that have a poor response to traditional therapies (eg. unmethylated MGMT). Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 5281. doi:10.1158/1538-7445.AM2011-5281