Default Configuration Research Articles

Bandwidth-Aware Dynamic Prefetch Configuration for IBM POWER8

Advanced hardware prefetch engines are being integrated in current high-performance processors. Prefetching can boost the performance of most applications, however, the induced bandwidth consumption can lead the system to a high contention for main memory bandwidth, which is a scarce resource in current multicores. In such a case, the system performance can be severely damaged. This article characterizes the applications’ behavior in an IBM POWER8 machine, which presents many prefetch settings, varying the bandwidth contention. The study reveals that the best prefetch setting for each application depends on the main memory bandwidth availability, that is, it depends on the co-running applications. Based on this study, we propose Bandwidth-Aware Prefetch Configuration (BAPC) a scalable adaptive prefetching algorithm that improves the performance of multi-program workloads. BAPC increases the performance of the applications in a 12, 15, and 16 percent of 6-, 8-, and 10-application workloads over the IBM POWER8 default configuration. In addition, BAPC reduces bandwidth consumption in 39, 42, and 45 percent, respectively.

Accurate Energy and Performance Prediction for Frequency-Scaled GPU Kernels

Energy optimization is an increasingly important aspect of today’s high-performance computing applications. In particular, dynamic voltage and frequency scaling (DVFS) has become a widely adopted solution to balance performance and energy consumption, and hardware vendors provide management libraries that allow the programmer to change both memory and core frequencies manually to minimize energy consumption while maximizing performance. This article focuses on modeling the energy consumption and speedup of GPU applications while using different frequency configurations. The task is not straightforward, because of the large set of possible and uniformly distributed configurations and because of the multi-objective nature of the problem, which minimizes energy consumption and maximizes performance. This article proposes a machine learning-based method to predict the best core and memory frequency configurations on GPUs for an input OpenCL kernel. The method is based on two models for speedup and normalized energy predictions over the default frequency configuration. Those are later combined into a multi-objective approach that predicts a Pareto-set of frequency configurations. Results show that our approach is very accurate at predicting extema and the Pareto set, and finds frequency configurations that dominate the default configuration in either energy or performance.

Dose-volume Histograms Comparison From Two Different Beam Configurations In The Monte Carlo Simulation of 6MV Clinac2100 Using PRIMO Program

The Monte Carlo (MC) method is widely considered to provide the most accurate dose distribution in external beam radiotherapy. A self-contained, full MC linac simulator and dose calculator, PRIMO is a windows-based, freely-distributed software which allows the importation of external compliant phase-space files among its many capabilities. Using PRIMO, a Varian Clinac2100 is simulated at 6 MV nominal energy for 108 number of histories. The dose calculations are compared using two different initial electron beam configurations. The tuned-beam profile has initial electron beam energy of 6.26 MeV, 0.150 MeV full-width-half-maximum (FWHM), 0.150 cm focal spot FWHM and 2 degrees beam divergence. The default configuration given in PRIMO has 5.40 MeV as initial electron beam energy and zero values for all the other beam parameters. A brain computerized tomography (CT) volume is imported into PRIMO and using its contouring tools, these structures are delineated: a hypothetical grosstumor volume (GTV), the brain, the left and right eye lens. The CT volume is irradiated with two parallel-opposed fields of size 10 cm by 10 cm with the brain conformed using the 52-leaf MLC. The dose-volume histogram (DVH) of the GTV and the brain for the tuned-beam profile has a larger extent into the high dose region compared to the DVHs of the default beam configuration. The dose given to 95% of the volume (D95) and the percentage of the volume receiving a dose equal to 95% (V95) are also compared. The GTV for the default beam profile has D95 = 86.31% with V95 = 0.44%, while for the tuned-beam profile, D95 = 88.30% with V95 = 0.95%. For the brain, D95 = 2.55% with V95 = 1.13% for the default profile while the tuned beam profile gives D95 = 2.43% with V95 = 2.33%. For the left [right] lens, D95 = 0.15% [D95 = 0.11%] for the default profile and D95 = 0.16% [D95 = 0.081%] for the tuned-beam profile. Both beam profiles give V95 = 0% for the left- and right- eye lens.

Optimized Approach (SPCA) for Load Balancing in Distributed HDFS Cluster

Hadoop is an open-source utility which allows users to provide massive input in terms of data and facilitates the computation. Role of Hadoop in load balancing is enormous which allows the user to configure the network of nodes having master/slave nodes. Hadoop’s typical architecture takes into consideration the default configuration for the machine as homogeneous, but many of the real-time application or clusters of nodes will have the homogeneous configurations. Thus, an effort is made in this paper to consider the homogeneity of the nodes in clusters and build an efficient algorithm which does load balancing in an efficient way when compared with the default balancer of Hadoop which works well only on homogeneous nodes.

FAAL: a Feature-based Aligning ALgorithm

Abstract In this article I present FAAL, a new Feature-based Aligning ALgorithm that can be used for the alignment of phonetically transcribed lexical items according to the comparison of phonetic features. The algorithm can be run on any pair of phonetically transcribed words without requiring any specific setting or tuning, although various parameters of its implementation as .jar library can indeed be modified by the user, if needed. The structure and work-flow of the algorithm are described in the first part of the paper. In the second part FAAL is tested against previously proposed algorithms in a test case involving two different datasets. FAAL in its default configuration outperforms all of them. FAAL is distributed in the form of a .jar Java library. The description of this Java library and the details about the settings and configuration of its methods are provided on-line, in my github repository at https://github.com/MKilani/FAAL.

A Fast and Scalable Workflow for SNPs Detection in Genome Sequences Using Hadoop Map-Reduce.

Next generation sequencing (NGS) technologies produce a huge amount of biological data, which poses various issues such as requirements of high processing time and large memory. This research focuses on the detection of single nucleotide polymorphism (SNP) in genome sequences. Currently, SNPs detection algorithms face several issues, e.g., computational overhead cost, accuracy, and memory requirements. In this research, we propose a fast and scalable workflow that integrates Bowtie aligner with Hadoop based Heap SNP caller to improve the SNPs detection in genome sequences. The proposed workflow is validated through benchmark datasets obtained from publicly available web-portals, e.g., NCBI and DDBJ DRA. Extensive experiments have been performed and the results obtained are compared with Bowtie and BWA aligner in the alignment phase, while compared with GATK, FaSD, SparkGA, Halvade, and Heap in SNP calling phase. Experimental results analysis shows that the proposed workflow outperforms existing frameworks e.g., GATK, FaSD, Heap integrated with BWA and Bowtie aligners, SparkGA, and Halvade. The proposed framework achieved 22.46% more efficient F-score and 99.80% consistent accuracy on average. More, comparatively 0.21% mean higher accuracy is achieved. Moreover, SNP mining has also been performed to identify specific regions in genome sequences. All the frameworks are implemented with the default configuration of memory management. The observations show that all workflows have approximately same memory requirement. In the future, it is intended to graphically show the mined SNPs for user-friendly interaction, analyze and optimize the memory requirements as well.

Effect of the tangential NBI current drive on the stability of pressure and energetic particle driven MHD modes in LHD plasma

The aim of the present study is to analyze the stability of the pressure gradient driven modes (PM) and Alfvén eigenmodes (AE) in the large helical device (LHD) plasma if the rotational transform profile is modified by the current drive of the tangential neutral beam injectors (NBI). This study forms a basic search for optimized operation scenarios with reduced mode activity. The analysis is performed using the code FAR3d which solves the reduced MHD equations describing the linear evolution of the poloidal flux and the toroidal component of the vorticity in a full 3D system, coupled with equations for density and parallel velocity moments of the energetic particle (EP) species, including the effect of the acoustic modes. The Landau damping and resonant destabilization effects are added via the closure relation. On-axis and off-axis NBI current drive modifies the rotational transform which becomes strongly distorted as the intensity of the neutral beam current drive (NBCD) increases, leading to wider continuum gaps and modifying the magnetic shear. The simulations with on-axis NBI injection show that a counter (ctr-) NBCD in inward shifted and default configurations leads to a lower growth rate of the PM, although strong n = 1 and 2 AEs can be destabilized. For the outward shifted configurations, a co-NBCD improves the AEs stability but the PM are further destabilized if the co-NBCD intensity is 30 kA T−1. If the NBI injection is off-axis, the plasma stability is not significantly improved due to the further destabilization of the AE and energetic particle modes (EPM) in the middle and outer plasma region.

Hdconfigor: Automatically Tuning High Dimensional Configuration Parameters for Log Search Engines

Search engines are nowadays widely applied to store and analyze logs generated by large-scale distributed systems. To adapt to various workload scenarios, log search engines such as Elasticsearch usually expose a large number of performance-related configuration parameters. As manual configuring is time consuming and labor intensive, automatically tuning configuration parameters to optimize performance has been an urgent need. However, it is challenging because: 1) Due to the complex implementation, the relationship between performance and configuration parameters is difficult to model and thus the objective function is actually a black box; 2) In addition to application parameters, JVM and kernel parameters are also closely related to the performance and together they construct a high dimensional configuration space; 3) To iteratively search for the best configuration, a tool is necessary to automatically deploy the newly generated configuration and launch tests to measure the corresponding performance. To address these challenges, this paper designs and implements HDConfigor, an automatic holistic configuration parameter tuning tool for log search engines. In order to solve the high dimensional optimization problem, we propose a modified Random EMbedding Bayesian Optimization algorithm (mREMBO) in HDConfigor which is a black-box approach. Instead of directly using a black-box optimization algorithm such as Bayesian optimization (BO), mREMBO first generates a lower dimensional embedded space through introducing a random embedding matrix and then performs BO in this embedded space. Therefore, HDConfigor is able to find a competitive configuration automatically and quickly. We evaluate HDConfigor in an Elasticsearch cluster with different workload scenarios. Experimental results show that compared with the default configuration, the best relative median indexing results achieved by mREMBO can reach $2.07\times $ . In addition, under the same number of trials, mREMBO is able to find a configuration with at least a further 10.31% improvement in throughput compared to Random search, Simulated Annealing and BO.

Deep Forest Regression for Short-Term Load Forecasting of Power Systems

Deep neural networks of deep learning algorithms can be applied into regressions and classifications. While the regression performances and classification performances of the deep neural networks are depending on the hyper-parameters of the deep neural networks. To mitigate the adverse effect of the hyper-parameters for the deep learning algorithms, this paper proposes deep forest regression for the short-term load forecasting of power systems. Deep forest regression includes two procedures, i.e., multi-grained scanning procedure and cascade forest procedure. These two procedures can be effectively trained by two completely random forests and two random forests with the default configuration. Then, the deep forest regression is applied into the short-term load forecasting of power systems. The forecasting performances of deep forest regression are compared with that of numerous intelligent algorithms and conventional regression algorithms under the model with the data of previous 7-day, 21-day, and 40-day. Besides, the forecasting performances of deep forest regression with different parameters are compared. The numerical results show that the deep forest regression with default configured parameters can increase the accuracy of the short-term forecasting and mitigate the influences of the experiences for the configuration of the hyper-parameters of deep learning model.

ATCS: Auto-Tuning Configurations of Big Data Frameworks Based on Generative Adversarial Nets

Big data processing frameworks (e.g., Spark, Storm) have been extensively used for massive data processing in the industry. To improve the performance and robustness of these frameworks, developers provide users with highly-configurable parameters. Due to the high-dimensional parameter space and complicated interactions of parameters, manual tuning of parameters is time-consuming and ineffective. Building performance-predicting models for big data frameworks is challenging for several reasons: (1) the significant time required to collect training data and (2) the poor accuracy of the prediction model when training data are limited. To meet this challenge, we proposes an auto-tuning configuration parameters system (ATCS), a new auto-tuning approach based on Generative Adversarial Nets (GAN). ATCS can build a performance prediction model with less training data and without sacrificing model accuracy. Moreover, an optimized Genetic Algorithm (GA) is used in ATCS to explore the parameter space for optimum solutions. To prove the effectiveness of ATCS, we select five frequently-used workloads in Spark, each of which runs on five different sized data sets. The results demonstrate that ATCS improves the performance of five frequently-used Spark workloads compared to the default configurations. We achieved a performance increase of 3.5× on average, with a maximum of 6.9×. To obtain similar model accuracy, experiment results also demonstrate that the quantity of ATCS training data is only 6% of Deep Neural Network (DNN) data, 13% of Support Vector Machine (SVM) data, 18% of Decision Tree (DT) data. Moreover, compared to other machine learning models, the average performance increase of ATCS is 1.7× that of DNN, 1.6× that of SVM, 1.7× that of DT on the five typical Spark programs.

Impact of different compilers and build types on Geant4 simulation execution time

Experimental observations and advanced computer simulations in High Energy Physics (HEP) paved the way for the recent discoveries at the Large Hadron Collider (LHC) at CERN. Currently, Monte Carlo simulations account for a very significant amount of computational resources of the Worldwide LHC Computing Grid (WLCG). The current growth in available computing performance will not be enough to fulfill the expected demand for the forthcoming High Luminosity run (HL-LHC). More efficient simulation codes are therefore required. This study focuses on evaluating the impact of different build methods on the simulation execution time. The Geant4 toolkit, the standard simulation code for the LHC experiments, consists of a set of libraries which can be either dynamically or statically linked to the simulation executable. Dynamic libraries are currently the preferred build method. In this work, three versions of the GCC compiler, namely 4.8.5, 6.2.0 and 8.2.0 have been used. In addition, a comparison between four optimization levels (Os, O1, O2 and O3) has also been performed. Static builds for all the GCC versions considered, exhibit a reduction in execution times of about 10%. Switching to newer GCC version results in an average of 30% improvement in the execution time regardless of the build type. In particular, a static build with GCC 8.2.0 leads to an improvement of about 34% with respect to the default configuration (GCC 4.8.5, dynamic, O2). The different GCC optimization flags do not affect the execution times.

Application of Open-Source Big-Data Framework in Marine Information Processing

Gao, X.; Wang, H., and Li, X., 2019. Application of open-source big-data framework in marine information processing. In: Li, L.; Wan, X.; and Huang, X. (eds.), Recent Developments in Practices and Research on Coastal Regions: Transportation, Environment and Economy. Journal of Coastal Research, Special Issue No. 98, pp. 187–190. Coconut Creek (Florida), ISSN 0749-0208.Open-source big-data systems can use a variety of processing technologies. For workloads that only require batch processing, Hadoop, which is less time sensitive and less expensive than other solutions, would be a good choice. For workloads that only require stream processing, Storm can support a wider range of languages and achieve very low latency processing, but the default configuration can produce duplicate results and cannot guarantee order. Samza's tight integration with YARN and Kafka provides greater flexibility, easier-to-use multiteam usage, and simpler replication and state management. The most suitable solution depends mainly on the state of the data to be processed, the time required for processing, and the desired result. Specifically, using a full-featured solution or a solution that focuses primarily on a project requires a careful trade-off. As it matures and is widely accepted, similar issues need to be considered when evaluating any emerging and innovative solutions.

CyberSPL: A Framework for the Verification of Cybersecurity Policy Compliance of System Configurations Using Software Product Lines

Cybersecurity attacks affect the compliance of cybersecurity policies of the organisations. Such disadvantages may be due to the absence of security configurations or the use of default configuration values of software products and systems. The complexity in the configuration of products and systems is a known challenge in the software industry since it includes a wide range of parameters to be taken into account. In other contexts, the configuration problems are solved using Software Product Lines. This is the reason why in this article the framework Cybersecurity Software Product Line (CyberSPL) is proposed. CyberSPL is based on a methodology to design product lines to verify cybersecurity policies according to the possible configurations. The patterns to configure the systems related to the cybersecurity aspects are grouped by defining various feature models. The automated analysis of these models allows us to diagnose possible problems in the security configurations, reducing or avoiding them. As support for this proposal, a multi-user and multi-platform solution has been implemented, enabling setting a catalogue of public or private feature models. Moreover, analysis and reasoning mechanisms have been integrated to obtain all the configurations of a model, to detect if a configuration is valid or not, including the root cause of problems for a given configuration. For validating the proposal, a real scenario is proposed where a catalogue of four different feature models is presented. In this scenario, the models have been analysed, different configurations have been validated, and several configurations with problems have been diagnosed.

Performance analysis of deep learning workloads using roofline trajectories

Over the last decade, technologies derived from convolutional neural networks (CNNs) called Deep Learning applications, have revolutionized fields as diverse as cancer detection, self-driving cars, virtual assistants, etc. However, many users of such applications are not experts in Machine Learning itself. Consequently, there is limited knowledge among the community to run such applications in an optimized manner. The performance question for Deep Learning applications has typically been addressed by employing bespoke hardware (e.g., GPUs) better suited for such compute-intensive operations. However, such a degree of performance is only accessibly at increasingly high financial costs leaving only big corporations and governments with resources sufficient enough to employ them at a large scale. As a result, an average user is only left with access to commodity clusters with, in many cases, only CPUs as the sole processing element. For such users to make effective use of resources at their disposal, concerted efforts are necessary to figure out optimal hardware and software configurations. This study is one such step in this direction as we use the Roofline model to perform a systematic analysis of representative CNN models and identify opportunities for black box and application-aware optimizations. Using the findings from our study, we are able to obtain up to 3.5$$\times$$ speedup compared to vanilla TensorFlow with default configurations.

Automating the Configuration of MapReduce: A Reinforcement Learning Scheme

With the exponential growth of data and the high demand for the analysis of large datasets, the MapReduce framework has been widely utilized to process data in a timely, cost-effective manner. It is well-known that the performance of MapReduce is limited by its default configuration parameters, and there are a few research studies that have focused on finding the optimal configurations to improve the performance of the MapReduce framework. Recently, machine learning based approaches have been receiving more attention to be utilized to auto configure the MapReduce parameters to account for the dynamic nature of the applications. In this article, we propose and develop a reinforcement learning (RL)-based scheme, named RL-MRCONF, to automatically configure the MapReduce parameters. Specifically, we explore and experiment with two variations of RL-MRCONF; one variation is based on the traditional RL algorithm and the second is based on the deep RL algorithm. Results obtained from simulations show that the RL-MRCONF has the ability to successfully and effectively auto-configure the MapReduce parameters dynamically according to changes in job types and computing resources. Moreover, simulation results show our proposed RL-MRCONF scheme outperforms the traditional RL-based implementation. Using datasets provided by MR-Perf, simulation results show that our proposed scheme provides around 50% performance improvement in terms of execution time when compared with MapReduce using default settings.

Convectively Coupled Equatorial Wave Simulations Using the ECMWF IFS and the NOAA GFS Cumulus Convection Schemes in the NOAA GFS Model

Abstract There is a longstanding challenge in numerical weather and climate prediction to accurately model tropical wave variability, including convectively coupled equatorial waves (CCEWs) and the Madden–Julian oscillation. For subseasonal prediction, the European Centre for Medium-Range Weather Forecasts (ECMWF) Integrated Forecasting System (IFS) has been shown to be superior to the NOAA Global Forecast System (GFS) in simulating tropical variability, suggesting that the ECMWF model is better at simulating the interaction between cumulus convection and the large-scale tropical circulation. In this study, we experiment with the cumulus convection scheme of the ECMWF IFS in a research version of the GFS to understand which aspects of the IFS cumulus convection scheme outperform those of the GFS convection scheme in the tropics. We show that the IFS cumulus convection scheme produces significantly different tropical moisture and temperature tendency profiles from those simulated by the GFS convection scheme when it is coupled with other physics schemes in the GFS physics package. We show that a consistent treatment of the interaction between parameterized convective plumes in the GFS planetary boundary layer (PBL) and the IFS convection scheme is required for the GFS to replicate the tropical temperature and moisture profiles simulated by the IFS model. The GFS model with the IFS convection scheme, and the consistent treatment between the convection and PBL schemes, produces much more organized convection in the tropics, and generates tropical waves that propagate more coherently than the GFS in its default configuration due to better simulated interaction between low-level convergence and precipitation.

Application of convolutional neural networks featuring Bayesian optimization for landslide susceptibility assessment

This study developed a deep learning based technique for the assessment of landslide susceptibility through a one-dimensional convolutional network (1D-CNN) and Bayesian optimisation in Southern Yangyang Province, South Korea. A total of 219 slide inventories and 17 slide conditioning variables were obtained for modelling. The data showed a complex scenario. Some past slides have spread over steep lands, while others have spread through flat terrain. Random forest (RF) served to keep only important factors for further analysis as a pre-processing measure. To select CNN hyperparameters, Bayesian optimization was used. Three methods contributed to overcoming the overfitting issue owing to small training data in our research. The selection of key factors by RF helped first of all to reduce information dimensionality. Second, the CNN model with 1D convolutions was intended to considerably decrease the number of its parameters. Third, a high rate of drop-out (0.66) helped reduce the CNN parameters. Overall accuracy, area under the receiver operating characteristics curve (AUROC) and 5-fold cross-validation were used to evaluate the models. CNN performance was compared to ANN and SVM. CNN achieved the highest accuracy on testing dataset (83.11%) and AUROC (0.880, 0.893, using testing and 5-fold CV, respectively). Bayesian optimization enhanced CNN accuracy by~3% (compared with default configuration). CNN could outperform ANN and SVM owing to its complicated architecture and handling of spatial correlations through convolution and pooling operations. In complex situations where some variables make a non-linear contribution to the occurrence of landslides, the method suggested could thus help develop landslide susceptibility maps.

IoT Security Configurability with Security-by-Contract.

Cybersecurity is one of the biggest challenges in the Internet of Things (IoT) domain, as well as one of its most embarrassing failures. As a matter of fact, nowadays IoT devices still exhibit various shortcomings. For example, they lack secure default configurations and sufficient security configurability. They also lack rich behavioural descriptions, failing to list provided and required services. To answer this problem, we envision a future where IoT devices carry behavioural contracts and Fog nodes store network policies. One requirement is that contract consistency must be easy to prove. Moreover, contracts must be easy to verify against network policies. In this paper, we propose to combine the security-by-contract (S × C) paradigm with Fog computing to secure IoT devices. Following our previous work, first we formally define the pillars of our proposal. Then, by means of a running case study, we show that we can model communication flows and prevent information leaks. Last, we show that our contribution enables a holistic approach to IoT security, and that it can also prevent unexpected chains of events.

CELAH KEAMANAN KREDENSIAL WINDOWS PADA GOOGLE CHROME

An attack that can leak authentication credentials on a Windows operating system by using the SMB file sharing protocol on a Windows operating system is an ever-present problem. It's been exploited in many ways, but the only solution found is limited to local area networks. Jonathan Brossard and Hormazd Billimoria recently presented one study involving internet attacks at the Black Hat security conference in 2015. However, no SMB-related attacks were published. In this paper, we will explain how an attack can cause Windows credential theft, which may affect the default configuration in Google Chrome browser.Key words— Kredensial Windows, Google Chrome, Vulnerability

Metode Optimasi Memcached sebagai NoSQL Key-value Memory Cache

Memcached is an application that is used to store client query results on the web into the memory server as a temporary storage (cache). The goal is that the web remains responsive even though many access the web. Memcached uses key-value and the LRU (Least Recenly Used) algorithm to store data. In the default configuration Memcached can handle web-based applications properly, but if it is faced with an actual situation, where the process of transferring data and cache objects swells to thousands to millions of items, optimization steps are needed so that Memcached services can always be optimal, not experiencing Input / Output (I / O) overhead, and low latency. In a review of this paper, we will show some of the latest research in memcached optimization efforts. Some methods that can be used are clustering are; Memory partitioning, Graphic Processor Unit hash, User Datagram Protocol (UDP) transmission, Solid State Drive Hybird Memory and Memcached Hadoop distributed File System (HDFS)Keywords : memcached, optimization, web-app, overhead, latency