Format:
1 online resource (268 pages)
ISBN:
9783030328139
Series Statement:
Lecture Notes in Computer Science Ser. v.11459
Note:
Description based on publisher supplied metadata and other sources
,
Intro -- BenchCouncil: Benchmarking and Promoting Innovative Techniques -- Organization -- Contents -- AI Benchmarking -- AIBench: Towards Scalable and Comprehensive Datacenter AI Benchmarking -- 1 Introduction -- 2 Related Work -- 3 Datacenter AI Benchmark Suite-AIBench -- 3.1 Datacenter AI Micro Benchmarks -- 3.2 Datacenter AI Component Benchmarks -- 3.3 Application Benchmarks -- 3.4 AI Competition -- 4 Conclusion -- References -- HPC AI500: A Benchmark Suite for HPC AI Systems -- 1 Introduction -- 2 Deep Learning in Scientific Computing -- 2.1 Extreme Weather Analysis -- 2.2 High Energy Physics -- 2.3 Cosmology -- 2.4 Summary -- 3 Benchmarking Methodology and Decisions -- 3.1 Methodology -- 3.2 The Selected Datasets -- 3.3 The Selected Workloads -- 3.4 Metrics -- 4 Reference Implementation -- 4.1 Component Benchmarks -- 4.2 Micro Benchmarks -- 5 Conclusion -- References -- Edge AIBench: Towards Comprehensive End-to-End Edge Computing Benchmarking -- 1 Introduction -- 2 Related Work -- 3 The Summary of Edge AIBench -- 3.1 ICU Patient Monitor -- 3.2 Surveillance Camera -- 3.3 Smart Home -- 3.4 Autonomous Vehicle -- 3.5 A Federated Learning Framework Testbed -- 4 Conclusion -- References -- AIoT Bench: Towards Comprehensive Benchmarking Mobile and Embedded Device Intelligence -- 1 Introduction -- 2 Benchmarking Requirements -- 3 AIoT Bench -- 4 Related Work -- 5 Conclusion -- References -- A Survey on Deep Learning Benchmarks: Do We Still Need New Ones? -- 1 Introduction -- 2 A Survey on Deep Learning Benchmarks -- 2.1 Stanford DAWNBench -- 2.2 Baidu DeepBench -- 2.3 Facebook AI Performance Evaluation Platform -- 2.4 ICT BigDataBench -- 2.5 Other Benchmarks -- 3 Discussion -- 3.1 Benchmark Comparison -- 3.2 Observations -- 4 Conclusion and Future Work -- References -- Cloud -- Benchmarking VM Startup Time in the Cloud -- 1 Introduction
,
2 Related Works -- 3 Methodology -- 3.1 Environment Setup -- 3.2 Algorithm -- 4 Result -- 4.1 By Instance Type -- 4.2 By Time of the Day -- 4.3 By Instance Location -- 4.4 By Cluster -- 5 Conclusions and Future Work -- References -- An Open Source Cloud-Based NoSQL and NewSQL Database Benchmarking Platform for IoT Data -- Abstract -- 1 Introduction -- 2 Background -- 2.1 NewSQL -- 2.2 NoSQL -- 2.3 MongoDB -- 2.4 VoltDB -- 2.5 Apache Kafka -- 2.6 Cloud Computing -- 3 Benchmarking Framework -- 3.1 Framework Components -- 3.2 Architecture -- 3.3 Data Generation and Consumption Algorithms -- 4 Experiments -- 4.1 System Configuration, Sensor Data Structure, and Formats -- 4.2 Experiment I: Data Injection with Different Volume and Velocity -- 4.3 Experiment II: Transactional Data Processing on High Volume of Data -- 4.4 Experiment III: Analytical Data Processing on High Volume of Data -- 4.5 Findings from Experiments -- 5 Related Work -- 6 Conclusion -- Acknowledgment -- References -- Scalability Evaluation of Big Data Processing Services in Clouds -- 1 Introduction -- 2 Related Work -- 2.1 Big Data Benchmarks -- 2.2 Scalability Evaluation of Big Data Processing Systems -- 3 Evaluation Model -- 4 Experiments -- 4.1 Experiment Environment -- 4.2 Scale-Out Analysis -- 4.3 Scale-Up Experiment -- 4.4 Experimental Results -- 5 Conclusion -- References -- PAIE: A Personal Activity Intelligence Estimator in the Cloud -- 1 Introduction -- 2 Related Work -- 3 Overview of PAIE -- 4 Statistic Issues -- 4.1 PAI Computing Mechanism -- 4.2 Statistical Modeling -- 4.3 PAI Estimating and Error Bounding -- 5 Implementing over Storm -- 6 Performance Evaluation -- 6.1 Experiment Methodology -- 6.2 Performance Analysis -- 6.3 Scalability Evaluation -- 7 Conclusions -- References -- DCMIX: Generating Mixed Workloads for the Cloud Data Center -- 1 Introduction -- 2 Related Work
,
3 DCMIX -- 3.1 Workloads -- 3.2 Mixed Workload Generator -- 4 System Entropy -- 5 Experiment and Experimental Analysis -- 5.1 Experimental Configurations and Methodology -- 5.2 Experiment Results and Observations -- 5.3 Summary -- 6 Conclusion -- References -- Machine-Learning Based Spark and Hadoop Workload Classification Using Container Performance Patterns -- 1 Introduction -- 1.1 Background -- 1.2 Problem -- 1.3 Limitations of Previous Approaches -- 1.4 Our Contribution -- 1.5 Resource Managers and Containers -- 2 Evaluation Methodology -- 2.1 Container Performance Metrics -- 2.2 Workloads and Workload Transitions -- 2.3 Parameter Settings -- 2.4 Hardware and Software -- 3 Results -- 3.1 Steady State Workload Characteristics -- 3.2 Dynamic Workload Characteristics - Workload Transitions -- 4 Identifying and Classifying Workloads -- 5 Detecting Workload Transitions -- 6 Relative Value and Importance of Container Performance Metrics -- 7 Conclusion -- References -- Testing Raft-Replicated Database Systems -- 1 Introduction -- 2 Background -- 2.1 Replicated State Machines -- 2.2 Raft Overview -- 3 System Model -- 4 Evaluation Metrics -- 4.1 Correctness -- 4.2 Performance -- 4.3 Scalability -- 5 Test Dimensions -- 5.1 Fault Type -- 5.2 Data Operation Type -- 5.3 System Configuration -- 6 Experiments -- 6.1 Experimental Setups -- 6.2 Recovery Time -- 6.3 Throughput and Latency -- 6.4 Stability -- 7 Related Work -- 8 Conclusion -- References -- Big Data -- Benchmarking for Transaction Processing Database Systems in Big Data Era -- 1 Introuduction -- 2 Requirements and Challenges -- 2.1 Data Generation -- 2.2 Workload Generation -- 2.3 Measurement Definition -- 2.4 Others -- 3 PeakBench: Benchmarking Transaction Processing Database Systems on Intensive Workloads -- 3.1 Business Description -- 3.2 Implementation of Benchmark Tool -- 3.3 Workloads
,
4 Test of PeakBench -- 5 Related Work -- 6 Conclusion -- References -- UMDISW: A Universal Multi-Domain Intelligent Scientific Workflow Framework for the Whole Life Cycle of Scientific Data -- 1 Introduction -- 2 The Status of UMDISW in System Architecture -- 3 The Model of UMDISW -- 3.1 Workflow and Task -- 3.2 Data Flow and Information Flow -- 3.3 Data Node and Algorithm Node -- 3.4 Example -- 4 The Structure and Execution of UMDISW -- 4.1 Running Service Layer -- 4.2 Workflow Execution Layer -- 4.3 Data Resource Layer -- 5 The Application Scenario of UMDISW -- 5.1 Fully Automated Workflow -- 5.2 Semi-custom Workflow -- 5.3 Fully Custom Workflow -- 6 The Implementation of UMDISW -- 7 Conclusion -- References -- MiDBench: Multimodel Industrial Big Data Benchmark -- 1 Introduction -- 2 Big Data Benchmarking Requirements -- 3 Related Work -- 4 Our Benchmarking Methodology -- 4.1 BoM Data Scenario Analysis -- 4.2 Analysis of Time Series Data Scenario -- 4.3 Unstructured Data Scenario Analysis -- 5 Synthetic Data Generation -- 6 Workload Characterization Experiments -- 6.1 Performance Tests on BoM Database Systems -- 6.2 Performance Tests on Time Series Database Systems -- 6.3 Performance Tests on Unstructured Database Systems -- 7 Conclusion -- References -- Modelling and Prediction -- Power Characterization of Memory Intensive Applications: Analysis and Implications -- 1 Motivation -- 2 Evolution of Server Energy Efficiency -- 2.1 Metrics of Energy Efficiency and Energy Proportionality -- 2.2 Experiment Setup -- 3 Experiment Results and Observations -- 3.1 Results of SPECpower Workload -- 3.2 Results of STREAM Workload -- 3.3 Insights on Energy Efficiency of Memory Intensive Applications -- 4 Related Work -- 5 Conclusions -- References -- Multi-USVs Coordinated Detection in Marine Environment with Deep Reinforcement Learning -- 1 Introduction
,
2 Background -- 2.1 USV Overview -- 2.2 Reinforcement Learning -- 3 Approach -- 3.1 Single-USV RL -- 3.2 Multi-USVs Coordinated Detection -- 4 Results and Discussion -- 5 Conclusion -- References -- EC-Bench: Benchmarking Onload and Offload Erasure Coders on Modern Hardware Architectures -- 1 Introduction -- 2 Background -- 2.1 Erasure Coding -- 2.2 Onload and Offload Erasure Coders -- 3 EC-Bench Design -- 3.1 Design -- 3.2 Parameter Space -- 3.3 Metrics -- 4 Evaluation -- 4.1 Open Source Libraries -- 4.2 Experimental Setup -- 4.3 Experimental Results -- 5 Related Work -- 6 Conclusion -- References -- Algorithm and Implementations -- Benchmarking SpMV Methods on Many-Core Platforms -- 1 Introduction -- 2 Benchmarking Methodology -- 2.1 Selected SpMV Methods -- 2.2 Selected Features -- 2.3 Hardware Configuration -- 2.4 Data Sets -- 2.5 Experimental Methods -- 3 Experimental Results -- 3.1 SpMV Performance -- 3.2 Best-Method Analysis -- 3.3 Correlation Analysis of Performance and Sparse Pattern -- 4 Conclusion -- References -- Benchmarking Parallel K-Means Cloud Type Clustering from Satellite Data -- 1 Introduction -- 2 Background -- 2.1 Cloud Joint Histograms -- 2.2 K-means Clustering -- 3 Implementation Details -- 3.1 OpenMP Based Implementation -- 3.2 OpenMP and MPI Based Implementation -- 3.3 Spark Based Implementation -- 4 Results -- 4.1 Code Validity -- 4.2 Performance -- 4.3 Cross Comparison -- 5 Related Work -- 6 Conclusions -- References -- Correction to: MiDBench: Multimodel Industrial Big Data Benchmark -- Correction to: Chapter "MiDBench: Multimodel Industrial Big Data Benchmark" in: C. Zheng and J. Zhan (Eds.): Benchmarking, Measuring, and Optimizing, LNCS 11459, https://doi.org/10.1007/978-3-030-32813-9_15 -- Author Index
Additional Edition:
Erscheint auch als Druck-Ausgabe Zheng, Chen Benchmarking, Measuring, and Optimizing Cham : Springer International Publishing AG,c2019 ISBN 9783030328122
Language:
English
Subjects:
Computer Science
Keywords:
Informationstechnik
;
Benchmarking
;
Cloud Computing
;
Big Data
;
Konferenzschrift
Bookmarklink
