Abstract

Purpose: To model the modifiable risk factors by using the logistic regression (LR) and artificial neural network (ANN) models for prediction of progression of Age-related Macular Degeneration (AMD) and cross-validate these models for their predictive accuracies in a population in South India. Methods: The data (N= 3,723) were analyzed from Andhra Pradesh Eye Disease Study (APEDS) on participants aged≥ 40 years. Sub-population data from this sample were drawn by using Random under Sampling (RUS)(n= 213) and combination of RUS and Random over Sampling (ROS)(n= 1420) techniques. The modifiable and non-modifiable risk factors which were elicited as part of the study were used to derive the LR based risk score models and the model fit was assessed using bootstrap method for internal validity. The ANN model was built for three sets of data using the multi-layer feed-forward back propagation …

Abstract

Hybrid programming model is becoming increasingly popular for HPC applications as it has the dual-advantage of exploiting inter-node distributed-memory scalability and intra-node shared-memory performance in a cluster system. One of the key challenges for energy efficient execution of hybrid programs is to determine time and energy efficient hardware configurations among a large system configuration space. Given a hybrid program with an execution time deadline and an energy budget, we propose a measurement-based analytical modelling approach to determine these system configurations. In contrast to current approaches, we model both inter and intra-node resource overlaps, memory contention among cores within a node and network contention across multiple nodes. The model invalidated against direct measurement using five representative HPC applications on Intel Xeon and ARM clusters having diverse time-energy performance. We show that a Pareto frontier consisting of optimal configurations exist for a hybrid program running on homogeneous clusters. To further optimize the Pareto frontier, we introduce a new metric, useful computation ratio (UCR) to quantify the degree of resource contentions and communication overheads in an execution. We discuss how UCR and Pareto-optimal configurations can be used in conjunction by system's designers to gain further insights into system resource imbalances, and how application developers can further fine-tune their hybrid programs

Abstract

Motivated by the explosion of Big Data analytics, performance improvements in low-power (wimpy) systems and the increasing energy efficiency of GPUs, this paper presents a time–energy performance analysis of MapReduce on heterogeneous systems with GPUs. We evaluate the time and energy performance of three MapReduce applications with diverse resource demands on a Hadoop–CUDA framework. As executing these applications on heterogeneous systems with GPUs is challenging, we introduce a novel lazy processing technique which requires no modifications to the underlying Hadoop framework. To analyze the impact of heterogeneity, we compare the heterogeneous CPU+GPU with the homogeneous CPU-only execution across three systems with diverse characteristics, (i) a traditional high-performance (brawny) Intel i7 system hosting a discrete 640-core Nvidia GPU of the latest Maxwell generation, (ii) a wimpy platform consisting of a quad-core ARM Cortex-A9 hosting the same discrete Maxwell GPU, and (iii) a wimpy platform integrating four ARM Cortex-A15 cores and 192 Nvidia Kepler GPU cores on the same chip. These systems encompass both intra-node heterogeneity with discrete GPUs and intra-chip heterogeneity with integrated GPUs. Our measurement-based performance analysis highlights the following results. For compute-intensive workloads, the brawny heterogeneous system achieves speedups of up to 2.3 and reduces the energy usage by almost half compared to the brawny homogeneous system. As expected, for applications where data transfers dominate the execution time, heterogeneity exhibits worse time–energy performance compared to homogeneous systems. For such applications, the heterogeneous wimpy A9 system with discrete GPU uses around 14 times the energy of homogeneous A9 system due to both system resource imbalances and high power overhead of the discrete GPU. However, comparing among heterogeneous systems, the wimpy A15 with integrated GPU uses the lowest energy across all workloads. This allows us to establish an execution time equivalence ratio between a single brawny node and multiple wimpy nodes. Based on this equivalence ratio, the wimpy nodes exhibit energy savings of two-thirds while maintaining the same execution time. This result advocates the potential usage of heterogeneous wimpy systems with integrated GPUs for Big Data analytics.

Abstract

In this work we show that given a set S of n points with coordinates on an n × n grid, we can construct data structures for (i) reporting and (ii) counting the maximal points in an axes-parallel query rectangle in sub-logarithmic time. We assume our model of computation to be the word RAM with size of each word being log n bits.

Abstract

In extant quantum secret sharing protocols, once the secret is shared in a quantum network (qnet) it cannot be retrieved, even if the dealer wishes that his/her secret no longer be available in the network. For instance, if the dealer is part of two qnets, say Q1 and Q2 and he/she subsequently finds that Q2 is more reliable than Q1, he/she may wish to transfer all her secrets from Q1 to Q2. Known protocols are inadequate to address such a revocation. In this work we address this problem by designing a protocol that enables the source/dealer to bring back the information shared in the network, if desired. Unlike classical revocation, the no-cloning theorem automatically ensures that the secret is no longer shared in the network.

Abstract

Software engineering researchers and their industrial counterparts have emphasized that research is essential for innovation. In practice, rather than ending up in a win-win situation, most of the times both the parties lose because of differences in expectations between the two sides. In this report, we briefly summarize the key insights from 1st International Workshop on Software Engineering Research and Industrial Practices (SER&IPs 2014) co-located with 36th International Conference on Software Engineering (ICSE 2014). The core lesson that stems out of this workshop is a definite need to have multiple approaches to deal with software engineering research-industrial practices partnership. Specifically, the workshop presented two keynotes: one from an industrial researcher explaining the need for academic expertise and an academic researcher on how their research added value to industry. An invited …

Abstract

The principal objective of modelling symposium is to help form an active community working on various aspects of modelling and its applications. The focus is on problems that would be critical 3-5 years out in future. For example, problem spaces such as enterprises, systems, cyber-physical systems, etc. will be given more importance. The event has keynote and 6 invited talks delivered by personnel from Academia and Industry around the world. ModSym invited students and researchers to submit posters and extended abstracts.

Abstract

We present a mathematical analysis and comparison of the performance of integrated on-chip semiconductor ring laser gyroscope (SRLG) fabricated using GaAs/AlGaAs and InP/InGaAsP technologies. The performance parameters of the gyro are modeled in terms of fundamental material, waveguide, and resonator parameters. In addition to this, influence of phenomena specific to semiconductor lasers such as nonlinear coupling, spatial hole burning, gain grating formation, and carrier induced index change on the gyro performance is also included. The analysis helps in identifying critical parameters, which must be optimized to improve the gyro performance.Best achievable performance of integrated SRLG is calculated, and design modifications are suggested to enhance it for high-performance military applications.

Abstract

We quantify the amount of correlations generated between two different output modes in the process of imperfect cloning and deletion processes. We use three different measures of quantum correlations and investigate their role in determining the fidelity of the cloning and the deletion. We obtain a bound on the total correlation generated in the successive processes of cloning and deleting operations. This bound displays a new kind of complementary relationship between the quantum correlations required in generating a copy of a quantum state and the amount of correlations required in bringing it back to the original state by deleting and vice versa. Our result shows that the better we clone (delete) a state, the more difficult it will be to bring the state back to its original form by the process of deleting (cloning).

Abstract

It has been suggested that there may exist quantum correlations that go beyond entanglement. The existence of such correlations can be revealed by quantum discord, but not by the conventional measure of entanglement. We argue that a state displays quantumness that can be of local and nonlocal origin. The physical quantity such as the quantum discord probes not only the nonlocal quantumness but also the local quantumness, such as the “local superposition”. This can be a reason why such measures are non-zero when there is no entanglement. We consider a generalized version of the Werner state to demonstrate the interplay of local quantumness, nonlocal quantumness, and classical mixedness of a state.