2018 International Conference on High Performance Computing & Simulation (HPCS)最新文献_第2页

OpenCL Performance Prediction using Architecture-Independent Features 使用与体系结构无关的特性的OpenCL性能预测

2018 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2018-07-01 DOI: 10.1109/HPCS.2018.00095

Beau Johnston, G. Falzon, Josh Milthorpe

{"title":"OpenCL Performance Prediction using Architecture-Independent Features","authors":"Beau Johnston, G. Falzon, Josh Milthorpe","doi":"10.1109/HPCS.2018.00095","DOIUrl":"https://doi.org/10.1109/HPCS.2018.00095","url":null,"abstract":"OpenCL is an attractive programming model for heterogeneous high-performance computing systems, with wide support from hardware vendors and significant performance portability. To support efficient scheduling on HPC systems it is necessary to perform accurate performance predictions for OpenCL workloads on varied compute devices, which is challenging due to diverse computation, communication and memory access characteristics which result in varying performance between devices. The Architecture Independent Workload Characterization (AIWC) tool can be used to characterize OpenCL kernels according to a set of architecture-independent features. This work presents a methodology where AIWC features are used to form a model capable of predicting accelerator execution times. We used this methodology to predict execution times for a set of 37 computational kernels running on 15 different devices representing a broad range of CPU, GPU and MIC architectures. The predictions are highly accurate, differing from the measured experimental run-times by an average of only 1.2%, and correspond to actual execution time mispredictions of 9 ps to 1 sec according to problem size. A previously unencountered code can be instrumented once and the AIWC metrics embedded in the kernel, to allow performance prediction across the full range of modelled devices. The results suggest that this methodology supports correct selection of the most appropriate device for a previously unen- countered code, which is highly relevant to the HPC scheduling setting.","PeriodicalId":308138,"journal":{"name":"2018 International Conference on High Performance Computing & Simulation (HPCS)","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124099753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 8

Snow Depth Retrieval Algorithm from Radar Backscattering Measurements at L- and X- Band Using Multi-Incidence Angles 基于多入射角的L波段和X波段雷达后向散射雪深反演算法

2018 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2018-07-01 DOI: 10.1109/HPCS.2018.00021

F. Mazeh, Bilal Hammoud, H. Ayad, F. Ndagijimana, G. Faour, M. Fadlallah, J. Jomaah

引用次数: 0

The NAS Benchmark Kernels for Single and Multi-Tenant Cloud Instances with LXC/KVM 使用LXC/KVM的单租户和多租户云实例的NAS基准内核

2018 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2018-07-01 DOI: 10.1109/HPCS.2018.00066

Anderson M. Maliszewski, Dalvan Griebler, C. Schepke, Alexander Ditter, D. Fey, L. G. Fernandes

{"title":"The NAS Benchmark Kernels for Single and Multi-Tenant Cloud Instances with LXC/KVM","authors":"Anderson M. Maliszewski, Dalvan Griebler, C. Schepke, Alexander Ditter, D. Fey, L. G. Fernandes","doi":"10.1109/HPCS.2018.00066","DOIUrl":"https://doi.org/10.1109/HPCS.2018.00066","url":null,"abstract":"Private IaaS clouds are an attractive environment for scientific workloads and applications. It provides advantages such as almost instantaneous availability of high-performance computing in a single node as well as compute clusters, easy access for researchers, and users that do not have access to conventional supercomputers. Furthermore, a cloud infrastructure provides elasticity and scalability to ensure and manage any software dependency on the system with no third-party dependency for researchers. However, one of the biggest challenges is to avoid significant performance degradation when migrating these applications from physical nodes to a cloud environment. Also, we lack more research investigations for multi-tenant cloud instances. In this paper, our goal is to perform a comparative performance evaluation of scientific applications with single and multi-tenancy cloud instances using KVM and LXC virtualization technologies under private cloud conditions. All analyses and evaluations were carried out based on NAS Benchmark kernels to simulate different types of workloads. We applied statistic significance tests to highlight the differences. The results have shown that applications running on LXC-based cloud instances outperform KVM-based cloud instances in 93.75% of the experiments w.r.t single tenant. Regarding multi-tenant, LXC instances outperform KVM instances in 45% of the results, where the performance differences were not as significant as expected.","PeriodicalId":308138,"journal":{"name":"2018 International Conference on High Performance Computing & Simulation (HPCS)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115807160","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 7

Examining Energy Efficiency of Vectorization Techniques Using a Gaussian Elimination 利用高斯消去法检验矢量化技术的能量效率

2018 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2018-07-01 DOI: 10.1109/HPCS.2018.00054

T. Jakobs, G. Rünger

引用次数: 7

Data Prefetching on In-order Processors 顺序处理器上的数据预取

2018 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2018-07-01 DOI: 10.1109/HPCS.2018.00061

Cristobal Ortega, Victor Garcia, Miquel Moretó, Marc Casas, Roxana Rusitoru

{"title":"Data Prefetching on In-order Processors","authors":"Cristobal Ortega, Victor Garcia, Miquel Moretó, Marc Casas, Roxana Rusitoru","doi":"10.1109/HPCS.2018.00061","DOIUrl":"https://doi.org/10.1109/HPCS.2018.00061","url":null,"abstract":"Low-power processors have attracted attention due to their energy-efficiency. A large market, such as the mobile one, relies on these processors for this very reason. Even High Performance Computing (HPC) systems are starting to consider low-power processors as a way to achieve exascale performance within 20MW, however, they must meet the right performance/Watt balance. Current low-power processors contain in-order cores, which cannot re-order instructions to avoid data dependency-induced stalls. Whilst this is useful to reduce the chip's total power consumption, it brings several challenges. Due to the evolving performance gap between memory and processor, memory is a significant bottleneck. In-order cores cannot re-order instructions and are memory latency bound, something data prefetching can help alleviate by ensuring data is readily available. In this work, we do an exhaustive analysis of available data prefetching techniques in state-of-the-art in-order cores. We analyze 5 static prefetchers and 2 dynamic aggressiveness and destination mechanisms applied to 3 data prefetchers on a set of HPC mini- and proxy-applications, whilst running on in-order processors. We show that next-line prefetching can achieve nearly top performance with a reasonable bandwidth consumption when throttled, whilst neighbor prefetchers have been found to be best, overall.","PeriodicalId":308138,"journal":{"name":"2018 International Conference on High Performance Computing & Simulation (HPCS)","volume":"48 33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132332905","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 4

Roofline Scaling Trajectories: A Method for Parallel Application and Architectural Performance Analysis 屋顶线缩放轨迹:一种并行应用和建筑性能分析方法

2018 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2018-07-01 DOI: 10.1109/HPCS.2018.00065

K. Ibrahim, Samuel Williams, L. Oliker

{"title":"Roofline Scaling Trajectories: A Method for Parallel Application and Architectural Performance Analysis","authors":"K. Ibrahim, Samuel Williams, L. Oliker","doi":"10.1109/HPCS.2018.00065","DOIUrl":"https://doi.org/10.1109/HPCS.2018.00065","url":null,"abstract":"The end of Dennard scaling signaled a shift in HPC supercomputer architectures from systems built from single- core processor architectures to systems built from multicore and eventually manycore architectures. This transition substantially complicated performance optimization and analysis as new programming models were created, new scaling methodologies deployed, and on-chip contention became a bottleneck to performance. Existing distributed memory performance models like logP and logGP were unable to capture this contention. The Roofline model was created to address this contention and its interplay with locality. However, to date, the Roofline model has focused on full-node concurrency. In this paper, we extend the Roofline model to capture the effects of concurrency on data locality and on-chip contention. We demonstrate the value of this new technique by evaluating the NAS parallel benchmarks on both multicore and manycore architectures under both strong-and weak-scaling regimes. In order to quantify the interplay between programming model and locality, we evaluate scaling under both the OpenMP and flat MPI programming models.","PeriodicalId":308138,"journal":{"name":"2018 International Conference on High Performance Computing & Simulation (HPCS)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133622198","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 8

Towards Probabilistic Networks of Polarized Evolutionary Processors 论极化进化处理器的概率网络

2018 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2018-07-01 DOI: 10.1109/HPCS.2018.00123

F. Arroyo, Sandra Gómez Canaval, V. Mitrana, M. Păun, José-Ramón Sánchez-Couso

{"title":"Towards Probabilistic Networks of Polarized Evolutionary Processors","authors":"F. Arroyo, Sandra Gómez Canaval, V. Mitrana, M. Păun, José-Ramón Sánchez-Couso","doi":"10.1109/HPCS.2018.00123","DOIUrl":"https://doi.org/10.1109/HPCS.2018.00123","url":null,"abstract":"The aim of this paper is to discuss two possible ways of introducing some features based on probabilistic concepts and methods in networks of polarized evolutionary processors (NPEP). We associate probabilities with rules in every node such that together with the communication protocol, which is based on the compatibility between the polarization of each node and data navigating through the network, might facilitate the study of biological phenomena as well as software simulations or hardware implementations. The probability associated with rules may be a priori defined and fixed or may be computed dynamically. Probabilities will also appear when communicating data between nodes; these probabilities may be statically or dynamically defined. This note also proposes the study of the impact of these characteristics and see how these new features reduce the gap between the formal model and its practical applicability. Introducing probabilities in NPEP is aimed to decrease the exponential expansion of the number of strings which appear in the computations used to solve NP-problems in a polynomial time. A decreasing of the exponential expansion of this number is achieved with a loss of certainty of the final result which is reached with some error probability.","PeriodicalId":308138,"journal":{"name":"2018 International Conference on High Performance Computing & Simulation (HPCS)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134464102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3

Interoperability Based Dynamic Data Mediation using Adaptive Multi-Agent Systems for Co-Simulation 基于互操作性的自适应多agent系统协同仿真动态数据中介

2018 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2018-07-01 DOI: 10.1109/HPCS.2018.00050

Yassine Motie, Elhadi Belghache, A. Nketsa, J. Georgé

引用次数: 2

Convolutional Neural Networks on Embedded Automotive Platforms: A Qualitative Comparison 嵌入式汽车平台上的卷积神经网络:定性比较

2018 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2018-07-01 DOI: 10.1109/HPCS.2018.00084

Gianluca Brilli, P. Burgio, M. Bertogna

引用次数: 5

Static Loop Parallelization Decision Using Template Metaprogramming 使用模板元编程的静态循环并行化决策

2018 International Conference on High Performance Computing & Simulation (HPCS) Pub Date : 2018-07-01 DOI: 10.1109/HPCS.2018.00159

Alexis Pereda, D. Hill, C. Mazel, Bruno Bachelet

引用次数: 0