High Performance Computing [[electronic resource] ] : ISC High Performance 2023 International Workshops, Hamburg, Germany, May 21–25, 2023, Revised Selected Papers / / edited by Amanda Bienz, Michèle Weiland, Marc Baboulin, Carola Kruse |
Autore | Bienz Amanda |
Edizione | [1st ed. 2023.] |
Pubbl/distr/stampa | Cham : , : Springer Nature Switzerland : , : Imprint : Springer, , 2023 |
Descrizione fisica | 1 online resource (677 pages) |
Disciplina |
621.39
004.6 |
Altri autori (Persone) |
WeilandMichèle
BaboulinMarc KruseCarola |
Collana | Lecture Notes in Computer Science |
Soggetto topico |
Computer engineering
Computer networks Computer Engineering and Networks |
ISBN | 3-031-40843-8 |
Formato | Materiale a stampa ![]() |
Livello bibliografico | Monografia |
Lingua di pubblicazione | eng |
Nota di contenuto | 2nd International Workshop on Malleability Techniques Applications in High-Performance Computing (HPCMALL) -- From Static to Malleable: Improving Flexibility and Compatibility in Burst Buffer File Systems -- Malleable techniques and resource scheduling to improve energy efficiency in parallel applications -- Towards Achieving Transparent Malleability Thanks to MPI Process Virtualization -- A Case Study on PMIx-usage for Dynamic Resource Management -- Malleable and adaptive ad-hoc file system for data intensive workloads in HPC applications -- Malleable and adaptive ad-hoc file system for data intensive workloads in HPC applications -- Towards Smarter Schedulers: Molding Jobs into the Right Shape via Monitoring and Modeling -- 18th Workshop on Virtualization in High-Performance Cloud Computing (VHPC 23) -- Improving live migration efficiency in QEMU: a paravirtualized approach -- Performance losses with virtualization: Comparing bare metal to VMs and containers -- Real-Time Unikernels: a First Look -- Accelerating Scientific Applications with the Quantum Edge: a Drug Design Use Case -- Event-Driven Chaos Testing For Containerized Applications -- HPC I/O in the Data Center (HPC IODC) -- Analyzing Parallel Applications for Unnecessary I/O Semantics That Inhibit File System Performance -- Workshop on Converged Computing of Cloud, HPC, and Edge (WOCC’23) -- Running Kubernetes Workloads on HPC -- A GPU-accelerated Molecular Docking Workflow with Kubernetes and Apache Airflow -- Cloud-Bursting and Autoscaling for Python-Native Scientific Workflows Using Ray -- Understanding System Resilience for Converged Computing of Cloud, Edge, and HPC -- Estimating the Energy Consumption of Applications in the Computing Continuum with iFogSim -- 7th International Workshop on In Situ Visualization (WOIV’23) -- Inshimtu – A Lightweight In Situ Visualization “Shim” -- Catalyst-ADIOS2: in transit analysis for numerical simulations using Catalyst 2 API -- A Case Study on Providing Accessibility-Focused In-Transit Architectures for Neural Network Simulation and Analysis -- Workshop on Monitoring and Operational Data Analytics (MODA23) -- Automatic Detection of HPC Job Inefficiencies at TU Dresden’s HPC center with PIKA -- ML-based methodology for HPC facilities supervision -- A Fast Simulator to Enable HPC Scheduling Strategy Comparisons -- 2nd Workshop on Communication, I/O, and Storage at Scale on Next-Generation Platforms: Scalable Infrastructures -- Application Performance Analysis: a Report on the Impact of Memory Bandwidth -- DAOS beyond Persistent Memory: Architecture and Initial Performance Results -- Enabling Multi-level Network Modeling in Structural Simulation Toolkit for Next-Generation HPC Network Design Space Exploration -- Portability and Scalability of OpenMP Offloading on State-of-the-art Accelerators -- An Earlier Experiences towards Optimizing Apache Spark over Frontera Supercomputer -- Bandwidth Limits in the Intel Xeon Max (Sapphire Rapids with HBM) Processors -- First International Workshop on RISC-V for HPC -- Test-driving RISC-V Vector hardware for HPC -- Backporting RISC-V Vector assembly -- Functional Testing with STLs: A Step Towards Reliable RISC-V-based HPC Commodity Clusters -- Challenges and Opportunities for RISC-V Arquitectures towards Genomics-based Workloads -- Optimizations for Very Long and Sparse Vector Operations on a RISC-V VPU : A Work-in-progress -- Performance Modelling-driven Optimization of RISC-V Hardware for Efficient SpMV -- Prototyping reconfigurable RRAM-based AI accelerators using the RISC-V ecosystem and Digital Twins -- Optimization of the FFT algorithm on RISC-V CPUs -- Software Development Vehicles to enable extended and early co-design: a RISC-V and HPC case of study -- Evaluation of HPC Workloads Running on Open-Source RISC-V Hardware -- Accelerating Neural Networks using Open Standard Software on RISC-V -- Second Combined Workshop on Interactive and Urgent Supercomputing (CWIUS) -- From Desktop to Supercomputer: Computational Fluid Dynamics Augmented by Molecular Dynamics using MaMiCo and preCICE -- Open OnDemand Connector for Amazon Elastic Kubernetes Service -- HPC on Heterogeneous Hardware (H3) -- GEMM-Like Convolution for Deep Learning Inference on the Xilinx Versal -- An Investigation into the Performance and Portability of SYCL Compiler Implementations -- Observed Memory Bandwidth and Power Usage on FPGA Platforms with oneAPI and Vitis HLS: A Comparison with GPUs -- Evaluating Quantum Algorithms for Linear Algebra Workflows -- Exploring the Use of Dataflow Architectures for Graph Neural Network Workloads -- OpenACC unified programming environment for multi-hybrid acceleration with GPU and FPGA. |
Record Nr. | UNISA-996546850503316 |
Bienz Amanda
![]() |
||
Cham : , : Springer Nature Switzerland : , : Imprint : Springer, , 2023 | ||
![]() | ||
Lo trovi qui: Univ. di Salerno | ||
|
High Performance Computing [[electronic resource] ] : ISC High Performance 2023 International Workshops, Hamburg, Germany, May 21–25, 2023, Revised Selected Papers / / edited by Amanda Bienz, Michèle Weiland, Marc Baboulin, Carola Kruse |
Autore | Bienz Amanda |
Edizione | [1st ed. 2023.] |
Pubbl/distr/stampa | Cham : , : Springer Nature Switzerland : , : Imprint : Springer, , 2023 |
Descrizione fisica | 1 online resource (677 pages) |
Disciplina |
621.39
004.6 |
Altri autori (Persone) |
WeilandMichèle
BaboulinMarc KruseCarola |
Collana | Lecture Notes in Computer Science |
Soggetto topico |
Computer engineering
Computer networks Computer Engineering and Networks |
ISBN | 3-031-40843-8 |
Formato | Materiale a stampa ![]() |
Livello bibliografico | Monografia |
Lingua di pubblicazione | eng |
Nota di contenuto | 2nd International Workshop on Malleability Techniques Applications in High-Performance Computing (HPCMALL) -- From Static to Malleable: Improving Flexibility and Compatibility in Burst Buffer File Systems -- Malleable techniques and resource scheduling to improve energy efficiency in parallel applications -- Towards Achieving Transparent Malleability Thanks to MPI Process Virtualization -- A Case Study on PMIx-usage for Dynamic Resource Management -- Malleable and adaptive ad-hoc file system for data intensive workloads in HPC applications -- Malleable and adaptive ad-hoc file system for data intensive workloads in HPC applications -- Towards Smarter Schedulers: Molding Jobs into the Right Shape via Monitoring and Modeling -- 18th Workshop on Virtualization in High-Performance Cloud Computing (VHPC 23) -- Improving live migration efficiency in QEMU: a paravirtualized approach -- Performance losses with virtualization: Comparing bare metal to VMs and containers -- Real-Time Unikernels: a First Look -- Accelerating Scientific Applications with the Quantum Edge: a Drug Design Use Case -- Event-Driven Chaos Testing For Containerized Applications -- HPC I/O in the Data Center (HPC IODC) -- Analyzing Parallel Applications for Unnecessary I/O Semantics That Inhibit File System Performance -- Workshop on Converged Computing of Cloud, HPC, and Edge (WOCC’23) -- Running Kubernetes Workloads on HPC -- A GPU-accelerated Molecular Docking Workflow with Kubernetes and Apache Airflow -- Cloud-Bursting and Autoscaling for Python-Native Scientific Workflows Using Ray -- Understanding System Resilience for Converged Computing of Cloud, Edge, and HPC -- Estimating the Energy Consumption of Applications in the Computing Continuum with iFogSim -- 7th International Workshop on In Situ Visualization (WOIV’23) -- Inshimtu – A Lightweight In Situ Visualization “Shim” -- Catalyst-ADIOS2: in transit analysis for numerical simulations using Catalyst 2 API -- A Case Study on Providing Accessibility-Focused In-Transit Architectures for Neural Network Simulation and Analysis -- Workshop on Monitoring and Operational Data Analytics (MODA23) -- Automatic Detection of HPC Job Inefficiencies at TU Dresden’s HPC center with PIKA -- ML-based methodology for HPC facilities supervision -- A Fast Simulator to Enable HPC Scheduling Strategy Comparisons -- 2nd Workshop on Communication, I/O, and Storage at Scale on Next-Generation Platforms: Scalable Infrastructures -- Application Performance Analysis: a Report on the Impact of Memory Bandwidth -- DAOS beyond Persistent Memory: Architecture and Initial Performance Results -- Enabling Multi-level Network Modeling in Structural Simulation Toolkit for Next-Generation HPC Network Design Space Exploration -- Portability and Scalability of OpenMP Offloading on State-of-the-art Accelerators -- An Earlier Experiences towards Optimizing Apache Spark over Frontera Supercomputer -- Bandwidth Limits in the Intel Xeon Max (Sapphire Rapids with HBM) Processors -- First International Workshop on RISC-V for HPC -- Test-driving RISC-V Vector hardware for HPC -- Backporting RISC-V Vector assembly -- Functional Testing with STLs: A Step Towards Reliable RISC-V-based HPC Commodity Clusters -- Challenges and Opportunities for RISC-V Arquitectures towards Genomics-based Workloads -- Optimizations for Very Long and Sparse Vector Operations on a RISC-V VPU : A Work-in-progress -- Performance Modelling-driven Optimization of RISC-V Hardware for Efficient SpMV -- Prototyping reconfigurable RRAM-based AI accelerators using the RISC-V ecosystem and Digital Twins -- Optimization of the FFT algorithm on RISC-V CPUs -- Software Development Vehicles to enable extended and early co-design: a RISC-V and HPC case of study -- Evaluation of HPC Workloads Running on Open-Source RISC-V Hardware -- Accelerating Neural Networks using Open Standard Software on RISC-V -- Second Combined Workshop on Interactive and Urgent Supercomputing (CWIUS) -- From Desktop to Supercomputer: Computational Fluid Dynamics Augmented by Molecular Dynamics using MaMiCo and preCICE -- Open OnDemand Connector for Amazon Elastic Kubernetes Service -- HPC on Heterogeneous Hardware (H3) -- GEMM-Like Convolution for Deep Learning Inference on the Xilinx Versal -- An Investigation into the Performance and Portability of SYCL Compiler Implementations -- Observed Memory Bandwidth and Power Usage on FPGA Platforms with oneAPI and Vitis HLS: A Comparison with GPUs -- Evaluating Quantum Algorithms for Linear Algebra Workflows -- Exploring the Use of Dataflow Architectures for Graph Neural Network Workloads -- OpenACC unified programming environment for multi-hybrid acceleration with GPU and FPGA. |
Record Nr. | UNINA-9910742497503321 |
Bienz Amanda
![]() |
||
Cham : , : Springer Nature Switzerland : , : Imprint : Springer, , 2023 | ||
![]() | ||
Lo trovi qui: Univ. Federico II | ||
|
High Performance Computing [[electronic resource] ] : ISC High Performance 2019 International Workshops, Frankfurt, Germany, June 16-20, 2019, Revised Selected Papers / / edited by Michèle Weiland, Guido Juckeland, Sadaf Alam, Heike Jagode |
Edizione | [1st ed. 2019.] |
Pubbl/distr/stampa | Cham : , : Springer International Publishing : , : Imprint : Springer, , 2019 |
Descrizione fisica | 1 online resource (XXV, 659 p. 402 illus., 239 illus. in color.) |
Disciplina | 004.3 |
Collana | Theoretical Computer Science and General Issues |
Soggetto topico |
Computer engineering
Computer networks Software engineering Computers Computer Engineering and Networks Software Engineering Computer Hardware Computing Milieux |
ISBN | 3-030-34356-1 |
Formato | Materiale a stampa ![]() |
Livello bibliografico | Monografia |
Lingua di pubblicazione | eng |
Nota di contenuto |
Intro -- Preface -- Organization -- Short Papers -- Preface to the First International Workshop on Legacy Software Refactoring for Performance -- P^3MA Workshop 2019 -- 4th International Workshop on In Situ Visualization (WOIV'19) -- Contents -- On the Use of Kernel Bypass Mechanisms for High-Performance Inter-container Communications -- 1 Introduction -- 2 Overview of Compared Solutions -- 3 Experimental Results -- 4 Related Work -- 5 Conclusions and Future Work -- References -- Continuous-Action Reinforcement Learning for Memory Allocation in Virtualized Servers -- 1 Introduction -- 2 Background -- 2.1 Memory Management in Virtualized Nodes -- 2.2 Reinforcement Learning: Markov Decision Process -- 3 CAVMem: Algorithm for Virtualized Memory Management -- 3.1 Decentralized Strategy for Memory Management -- 3.2 Formulating the Problem as an MDP -- 4 Experimental Framework -- 5 Results for Evaluation -- 5.1 Results for Scenario 1 -- 5.2 Results for Scenario 2 -- 5.3 Results for Scenario 3 -- 5.4 Discussion -- 6 Related Work -- 7 Conclusions and Future Work -- References -- Container Orchestration on HPC Clusters -- 1 Introduction -- 2 Related Work -- 3 Background -- 3.1 Kubernetes -- 3.2 Kubernetes Deployment -- 4 Implementation -- 4.1 General Approach -- 4.2 Kubernetes Cluster Deployment -- 4.3 HPC Worker Node Software Prerequisites -- 4.4 Networking -- 4.5 GE Worker Setup and Tear down -- 4.6 Kubernetes Cluster Configuration -- 5 Evaluation -- 6 Discussion -- 7 Conclusion and Future Work -- References -- Data Pallets: Containerizing Storage for Reproducibility and Traceability -- 1 Introduction -- 2 Related Work -- 3 Design -- 3.1 Design and Implementation Challenges -- 3.2 Design and Implementation Details -- 3.3 Integration with Sandia Analysis Workbench (SAW) -- 4 Measurements -- 4.1 Time Overheads -- 4.2 Space Overheads -- 4.3 Discussion.
5 Integration with Sandia Analysis Workbench -- 6 Conclusions and Future Work -- References -- Sarus: Highly Scalable Docker Containers for HPC Systems -- 1 Introduction -- 2 Related Work -- 3 Sarus -- 3.1 Sarus Architecture -- 3.2 Container Creation -- 4 Extending Sarus with OCI Hooks -- 4.1 Native MPICH-Based MPI Support (H1) -- 4.2 NVIDIA GPU Support (H2) -- 4.3 SSH Connection Within Containers (H3) -- 4.4 Slurm Scheduler Synchronization (H4) -- 5 Performance Evaluation -- 5.1 Scientific Applications -- 6 Conclusions -- References -- Singularity GPU Containers Execution on HPC Cluster -- 1 Introduction -- 2 Singularity GPU Containers Building and Running -- 3 Benchmark -- 3.1 Systems Description -- 3.2 Test Case 1: Containerized Tensorflow Execution on GALILEO Versus Official Tensorflow Performance Data -- 3.3 Test Case 2: Containerized Versus Bare Metal Execution on GALILEO -- 4 Conclusion -- References -- A Multitenant Container Platform with OKD, Harbor Registry and ELK -- 1 Introduction -- 2 Past -- 2.1 Background -- 2.2 Challenges -- 3 Present -- 3.1 Evaluation of Container Orchestration Frameworks -- 3.2 Observability: Logging and OKD -- 3.3 Observability: Monitoring and OKD -- 4 Future -- 4.1 Monitoring -- 4.2 Container Policy and OKD -- 4.3 Gitops gitops and OKD -- 4.4 Continuous Delivery in OKD -- 4.5 OKD in the Cloud -- 5 Conclusion -- References -- Enabling GPU-Enhanced Computer Vision and Machine Learning Research Using Containers -- 1 Introduction -- 2 Defining the Base Container -- 2.1 System Setup: Ubuntu, CUDA, Docker, Nvidia-Docker -- 2.2 Docker and Container Runtime -- 2.3 TensorFlow -- 2.4 OpenCV -- 2.5 Cuda_tensorflow_opencv -- 3 Using the Base Container -- 3.1 Testing Code from a Bash Terminal -- 3.2 Integrating Darknet and Yolo V3 Python Bindings -- 4 Conclusion -- References. Software and Hardware Co-design for Low-Power HPC Platforms -- 1 Introduction -- 2 Network Interface Primitives -- 3 HPC Prototype -- 4 User-Level Communication Library -- 5 MPI Implementation over the Proposed Architecture -- 6 Conclusions and Future Work -- References -- Modernizing Titan2D, a Parallel AMR Geophysical Flow Code to Support Multiple Rheologies and Extendability -- 1 Introduction -- 2 Titan2D and Benchmark Problem -- 3 Refactoring Strategies -- 3.1 Adopting a Python Interface -- 3.2 Merging Multiple Forks -- 3.3 Changing Data Layout to for Modern CPU Architectures -- 3.4 Efficient Indexing for Elements/Nodes Addressing -- 3.5 Introducing OpenMP and Hybrid OpenMP/MPI Parallelization -- 4 Performance Improvement Evaluation -- 5 Conclusions and Future Plans -- References -- Asynchronous AMR on Multi-GPUs -- 1 Introduction -- 2 Execution on Heterogeneous Architectures -- 2.1 Data Model and CPU-GPU Communication -- 2.2 Scheduling on Heterogeneous Architectures -- 2.3 API -- 2.4 Multi-GPU Support -- 3 Evaluation -- 4 Conclusions -- References -- Batch Solution of Small PDEs with the OPS DSL -- 1 Introduction -- 2 The OPS DSL -- 3 Batching Support in OPS -- 3.1 Extending the Abstraction -- 3.2 Execution Schedule Transformation -- 3.3 Data Layout Transformation -- 3.4 Alternating Direction Implicit Solver -- 4 Evaluation -- 4.1 The Application -- 4.2 Experimental Set-Up -- 4.3 Results -- 5 Conclusions -- References -- Scalable Parallelization of Stencils Using MODA -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 MODA and User-Defined Indices -- 3.2 Using GGDML Indices -- 3.3 Communication Identification -- 4 Evaluation -- 4.1 Test Application -- 4.2 Test System -- 4.3 Experiments -- 5 Summary -- References -- Comparing High Performance Computing Accelerator Programming Models -- 1 Introduction -- 2 Motivation -- 3 Related Work. 4 Analysis -- 5 Discussion -- 5.1 BT Benchmark -- 5.2 SP Benchmark -- 5.3 LBM Benchmark -- 5.4 LBDC Benchmark -- 6 Conclusion -- References -- Tracking User-Perceived I/O Slowdown via Probing -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 Probing -- 3.2 Data Reduction Using Statistics -- 3.3 Computing the Slowdown -- 4 Evaluation -- 4.1 Test Systems -- 4.2 Probing Tool -- 4.3 Timeseries of Individual Measurements -- 4.4 Host Variability -- 4.5 Understanding Application Behavior - The IO-500 -- 4.6 Long-Period -- 4.7 Slowdown -- 5 Conclusion -- References -- A Quantitative Approach to Architecting All-Flash Lustre File Systems -- 1 Introduction -- 2 Methods -- 3 File System Capacity -- 4 Drive Endurance -- 5 Metadata Configuration -- 5.1 MDT Capacity Required by DOM -- 5.2 MDT Capacity Required for Inodes -- 5.3 Overall MDT Capacity -- 6 Conclusion -- References -- MBWU: Benefit Quantification for Data Access Function Offloading -- 1 Introduction -- 2 The MBWU-Based Methodology -- 2.1 Background -- 2.2 What Is MBWU -- 2.3 How to Measure MBWU(s) -- 2.4 Evaluation Prototype -- 3 Evaluation -- 3.1 Infrastructure -- 3.2 Test Setup and Results -- 4 Related Work -- 5 Conclusion -- References -- Footprinting Parallel I/O - Machine Learning to Classify Application's I/O Behavior -- 1 Introduction -- 2 Related Work -- 3 DKRZ Monitoring -- 3.1 Metrics -- 4 Methodology -- 5 Test Data -- 5.1 Data Preparation -- 6 Evaluation -- 6.1 I/O Behavior Classification -- 6.2 Footprinting -- 7 Manual Identification of I/O Intensive Jobs -- 8 Summary and Conclusion -- References -- Adventures in NoSQL for Metadata Management -- 1 Introduction -- 2 Related Work -- 3 Metadata Model -- 3.1 Basic Metadata -- 3.2 Custom Metadata -- 4 Design -- 4.1 What Has the Right Features to Be Worth Testing? -- 4.2 What Is It Going to Take to Get It All Working at All?. 4.3 Can We Make Our Queries Work with Any Performance? -- 4.4 Battle Scars and Lessons for Our Next Battle Against Scale Out Computing Tools -- 5 Evaluation -- 5.1 Insert Time -- 5.2 Query Time -- 6 Conclusion and Future Work -- References -- Towards High Performance Data Analytics for Climate Change -- 1 Introduction -- 2 Main Challenges -- 3 The Ophidia Project -- 3.1 Multi-dimensional Storage Model -- 3.2 Array-Based Primitives and Parallel Operators -- 4 Benchmark and Experimental Results -- 4.1 Benchmark Definition -- 4.2 Test Environment -- 4.3 Experimental Results and Discussion -- 5 Related Work -- 6 Conclusions -- References -- An Architecture for High Performance Computing and Data Systems Using Byte-Addressable Persistent Memory -- 1 Introduction -- 2 Persistent Memory -- 2.1 Data Access -- 2.2 B-APM Modes of Operation -- 2.3 Non-volatile Memory Software Ecosystem -- 3 Opportunities for Exploiting B-APM for Computational Simulations and Data Analytics -- 3.1 Potential Caveats -- 4 Systemware Architecture -- 4.1 Job Scheduler -- 4.2 Data Scheduler -- 5 Performance Evaluation -- 6 Related Work -- 7 Summary -- References -- Mediating Data Center Storage Diversity in HPC Applications with FAODEL -- 1 Introduction -- 2 FAODEL Background -- 2.1 Kelpie -- 2.2 I/O Management (IOM) Modules -- 3 Mediating Storage Using Kelpie Object Naming -- 3.1 Kelpie Architectural Considerations -- 3.2 Annotating the Kelpie Namespace -- 3.3 Service-Initiated Mediation -- 3.4 Performance Considerations -- 4 Related Work -- 5 Conclusion -- References -- Predicting File Lifetimes with Machine Learning -- 1 Introduction -- 2 Specifying the Problem and Building the Models -- 2.1 Problem Specification -- 2.2 Dataset -- 2.3 Data Preprocessing -- 2.4 Models -- 3 Results -- 3.1 Evaluation Methodology -- 3.2 Training Times and Model Sizes -- 3.3 Accuracy. 3.4 Error and Accuracy Distribution. |
Record Nr. | UNISA-996466292803316 |
Cham : , : Springer International Publishing : , : Imprint : Springer, , 2019 | ||
![]() | ||
Lo trovi qui: Univ. di Salerno | ||
|
High Performance Computing [[electronic resource] ] : 34th International Conference, ISC High Performance 2019, Frankfurt/Main, Germany, June 16–20, 2019, Proceedings / / edited by Michèle Weiland, Guido Juckeland, Carsten Trinitis, Ponnuswamy Sadayappan |
Edizione | [1st ed. 2019.] |
Pubbl/distr/stampa | Cham : , : Springer International Publishing : , : Imprint : Springer, , 2019 |
Descrizione fisica | 1 online resource (XVI, 352 p. 512 illus., 113 illus. in color.) |
Disciplina | 004.3 |
Collana | Theoretical Computer Science and General Issues |
Soggetto topico |
Software engineering
Logic design Microprocessors Computer architecture Artificial intelligence Computer networks Software Engineering Logic Design Processor Architectures Artificial Intelligence Computer Communication Networks |
ISBN | 3-030-20656-4 |
Formato | Materiale a stampa ![]() |
Livello bibliografico | Monografia |
Lingua di pubblicazione | eng |
Record Nr. | UNISA-996466325603316 |
Cham : , : Springer International Publishing : , : Imprint : Springer, , 2019 | ||
![]() | ||
Lo trovi qui: Univ. di Salerno | ||
|
High Performance Computing [[electronic resource] ] : ISC High Performance 2019 International Workshops, Frankfurt, Germany, June 16-20, 2019, Revised Selected Papers / / edited by Michèle Weiland, Guido Juckeland, Sadaf Alam, Heike Jagode |
Edizione | [1st ed. 2019.] |
Pubbl/distr/stampa | Cham : , : Springer International Publishing : , : Imprint : Springer, , 2019 |
Descrizione fisica | 1 online resource (XXV, 659 p. 402 illus., 239 illus. in color.) |
Disciplina | 004.3 |
Collana | Theoretical Computer Science and General Issues |
Soggetto topico |
Computer engineering
Computer networks Software engineering Computers Computer Engineering and Networks Software Engineering Computer Hardware Computing Milieux |
ISBN | 3-030-34356-1 |
Formato | Materiale a stampa ![]() |
Livello bibliografico | Monografia |
Lingua di pubblicazione | eng |
Nota di contenuto |
Intro -- Preface -- Organization -- Short Papers -- Preface to the First International Workshop on Legacy Software Refactoring for Performance -- P^3MA Workshop 2019 -- 4th International Workshop on In Situ Visualization (WOIV'19) -- Contents -- On the Use of Kernel Bypass Mechanisms for High-Performance Inter-container Communications -- 1 Introduction -- 2 Overview of Compared Solutions -- 3 Experimental Results -- 4 Related Work -- 5 Conclusions and Future Work -- References -- Continuous-Action Reinforcement Learning for Memory Allocation in Virtualized Servers -- 1 Introduction -- 2 Background -- 2.1 Memory Management in Virtualized Nodes -- 2.2 Reinforcement Learning: Markov Decision Process -- 3 CAVMem: Algorithm for Virtualized Memory Management -- 3.1 Decentralized Strategy for Memory Management -- 3.2 Formulating the Problem as an MDP -- 4 Experimental Framework -- 5 Results for Evaluation -- 5.1 Results for Scenario 1 -- 5.2 Results for Scenario 2 -- 5.3 Results for Scenario 3 -- 5.4 Discussion -- 6 Related Work -- 7 Conclusions and Future Work -- References -- Container Orchestration on HPC Clusters -- 1 Introduction -- 2 Related Work -- 3 Background -- 3.1 Kubernetes -- 3.2 Kubernetes Deployment -- 4 Implementation -- 4.1 General Approach -- 4.2 Kubernetes Cluster Deployment -- 4.3 HPC Worker Node Software Prerequisites -- 4.4 Networking -- 4.5 GE Worker Setup and Tear down -- 4.6 Kubernetes Cluster Configuration -- 5 Evaluation -- 6 Discussion -- 7 Conclusion and Future Work -- References -- Data Pallets: Containerizing Storage for Reproducibility and Traceability -- 1 Introduction -- 2 Related Work -- 3 Design -- 3.1 Design and Implementation Challenges -- 3.2 Design and Implementation Details -- 3.3 Integration with Sandia Analysis Workbench (SAW) -- 4 Measurements -- 4.1 Time Overheads -- 4.2 Space Overheads -- 4.3 Discussion.
5 Integration with Sandia Analysis Workbench -- 6 Conclusions and Future Work -- References -- Sarus: Highly Scalable Docker Containers for HPC Systems -- 1 Introduction -- 2 Related Work -- 3 Sarus -- 3.1 Sarus Architecture -- 3.2 Container Creation -- 4 Extending Sarus with OCI Hooks -- 4.1 Native MPICH-Based MPI Support (H1) -- 4.2 NVIDIA GPU Support (H2) -- 4.3 SSH Connection Within Containers (H3) -- 4.4 Slurm Scheduler Synchronization (H4) -- 5 Performance Evaluation -- 5.1 Scientific Applications -- 6 Conclusions -- References -- Singularity GPU Containers Execution on HPC Cluster -- 1 Introduction -- 2 Singularity GPU Containers Building and Running -- 3 Benchmark -- 3.1 Systems Description -- 3.2 Test Case 1: Containerized Tensorflow Execution on GALILEO Versus Official Tensorflow Performance Data -- 3.3 Test Case 2: Containerized Versus Bare Metal Execution on GALILEO -- 4 Conclusion -- References -- A Multitenant Container Platform with OKD, Harbor Registry and ELK -- 1 Introduction -- 2 Past -- 2.1 Background -- 2.2 Challenges -- 3 Present -- 3.1 Evaluation of Container Orchestration Frameworks -- 3.2 Observability: Logging and OKD -- 3.3 Observability: Monitoring and OKD -- 4 Future -- 4.1 Monitoring -- 4.2 Container Policy and OKD -- 4.3 Gitops gitops and OKD -- 4.4 Continuous Delivery in OKD -- 4.5 OKD in the Cloud -- 5 Conclusion -- References -- Enabling GPU-Enhanced Computer Vision and Machine Learning Research Using Containers -- 1 Introduction -- 2 Defining the Base Container -- 2.1 System Setup: Ubuntu, CUDA, Docker, Nvidia-Docker -- 2.2 Docker and Container Runtime -- 2.3 TensorFlow -- 2.4 OpenCV -- 2.5 Cuda_tensorflow_opencv -- 3 Using the Base Container -- 3.1 Testing Code from a Bash Terminal -- 3.2 Integrating Darknet and Yolo V3 Python Bindings -- 4 Conclusion -- References. Software and Hardware Co-design for Low-Power HPC Platforms -- 1 Introduction -- 2 Network Interface Primitives -- 3 HPC Prototype -- 4 User-Level Communication Library -- 5 MPI Implementation over the Proposed Architecture -- 6 Conclusions and Future Work -- References -- Modernizing Titan2D, a Parallel AMR Geophysical Flow Code to Support Multiple Rheologies and Extendability -- 1 Introduction -- 2 Titan2D and Benchmark Problem -- 3 Refactoring Strategies -- 3.1 Adopting a Python Interface -- 3.2 Merging Multiple Forks -- 3.3 Changing Data Layout to for Modern CPU Architectures -- 3.4 Efficient Indexing for Elements/Nodes Addressing -- 3.5 Introducing OpenMP and Hybrid OpenMP/MPI Parallelization -- 4 Performance Improvement Evaluation -- 5 Conclusions and Future Plans -- References -- Asynchronous AMR on Multi-GPUs -- 1 Introduction -- 2 Execution on Heterogeneous Architectures -- 2.1 Data Model and CPU-GPU Communication -- 2.2 Scheduling on Heterogeneous Architectures -- 2.3 API -- 2.4 Multi-GPU Support -- 3 Evaluation -- 4 Conclusions -- References -- Batch Solution of Small PDEs with the OPS DSL -- 1 Introduction -- 2 The OPS DSL -- 3 Batching Support in OPS -- 3.1 Extending the Abstraction -- 3.2 Execution Schedule Transformation -- 3.3 Data Layout Transformation -- 3.4 Alternating Direction Implicit Solver -- 4 Evaluation -- 4.1 The Application -- 4.2 Experimental Set-Up -- 4.3 Results -- 5 Conclusions -- References -- Scalable Parallelization of Stencils Using MODA -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 MODA and User-Defined Indices -- 3.2 Using GGDML Indices -- 3.3 Communication Identification -- 4 Evaluation -- 4.1 Test Application -- 4.2 Test System -- 4.3 Experiments -- 5 Summary -- References -- Comparing High Performance Computing Accelerator Programming Models -- 1 Introduction -- 2 Motivation -- 3 Related Work. 4 Analysis -- 5 Discussion -- 5.1 BT Benchmark -- 5.2 SP Benchmark -- 5.3 LBM Benchmark -- 5.4 LBDC Benchmark -- 6 Conclusion -- References -- Tracking User-Perceived I/O Slowdown via Probing -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 Probing -- 3.2 Data Reduction Using Statistics -- 3.3 Computing the Slowdown -- 4 Evaluation -- 4.1 Test Systems -- 4.2 Probing Tool -- 4.3 Timeseries of Individual Measurements -- 4.4 Host Variability -- 4.5 Understanding Application Behavior - The IO-500 -- 4.6 Long-Period -- 4.7 Slowdown -- 5 Conclusion -- References -- A Quantitative Approach to Architecting All-Flash Lustre File Systems -- 1 Introduction -- 2 Methods -- 3 File System Capacity -- 4 Drive Endurance -- 5 Metadata Configuration -- 5.1 MDT Capacity Required by DOM -- 5.2 MDT Capacity Required for Inodes -- 5.3 Overall MDT Capacity -- 6 Conclusion -- References -- MBWU: Benefit Quantification for Data Access Function Offloading -- 1 Introduction -- 2 The MBWU-Based Methodology -- 2.1 Background -- 2.2 What Is MBWU -- 2.3 How to Measure MBWU(s) -- 2.4 Evaluation Prototype -- 3 Evaluation -- 3.1 Infrastructure -- 3.2 Test Setup and Results -- 4 Related Work -- 5 Conclusion -- References -- Footprinting Parallel I/O - Machine Learning to Classify Application's I/O Behavior -- 1 Introduction -- 2 Related Work -- 3 DKRZ Monitoring -- 3.1 Metrics -- 4 Methodology -- 5 Test Data -- 5.1 Data Preparation -- 6 Evaluation -- 6.1 I/O Behavior Classification -- 6.2 Footprinting -- 7 Manual Identification of I/O Intensive Jobs -- 8 Summary and Conclusion -- References -- Adventures in NoSQL for Metadata Management -- 1 Introduction -- 2 Related Work -- 3 Metadata Model -- 3.1 Basic Metadata -- 3.2 Custom Metadata -- 4 Design -- 4.1 What Has the Right Features to Be Worth Testing? -- 4.2 What Is It Going to Take to Get It All Working at All?. 4.3 Can We Make Our Queries Work with Any Performance? -- 4.4 Battle Scars and Lessons for Our Next Battle Against Scale Out Computing Tools -- 5 Evaluation -- 5.1 Insert Time -- 5.2 Query Time -- 6 Conclusion and Future Work -- References -- Towards High Performance Data Analytics for Climate Change -- 1 Introduction -- 2 Main Challenges -- 3 The Ophidia Project -- 3.1 Multi-dimensional Storage Model -- 3.2 Array-Based Primitives and Parallel Operators -- 4 Benchmark and Experimental Results -- 4.1 Benchmark Definition -- 4.2 Test Environment -- 4.3 Experimental Results and Discussion -- 5 Related Work -- 6 Conclusions -- References -- An Architecture for High Performance Computing and Data Systems Using Byte-Addressable Persistent Memory -- 1 Introduction -- 2 Persistent Memory -- 2.1 Data Access -- 2.2 B-APM Modes of Operation -- 2.3 Non-volatile Memory Software Ecosystem -- 3 Opportunities for Exploiting B-APM for Computational Simulations and Data Analytics -- 3.1 Potential Caveats -- 4 Systemware Architecture -- 4.1 Job Scheduler -- 4.2 Data Scheduler -- 5 Performance Evaluation -- 6 Related Work -- 7 Summary -- References -- Mediating Data Center Storage Diversity in HPC Applications with FAODEL -- 1 Introduction -- 2 FAODEL Background -- 2.1 Kelpie -- 2.2 I/O Management (IOM) Modules -- 3 Mediating Storage Using Kelpie Object Naming -- 3.1 Kelpie Architectural Considerations -- 3.2 Annotating the Kelpie Namespace -- 3.3 Service-Initiated Mediation -- 3.4 Performance Considerations -- 4 Related Work -- 5 Conclusion -- References -- Predicting File Lifetimes with Machine Learning -- 1 Introduction -- 2 Specifying the Problem and Building the Models -- 2.1 Problem Specification -- 2.2 Dataset -- 2.3 Data Preprocessing -- 2.4 Models -- 3 Results -- 3.1 Evaluation Methodology -- 3.2 Training Times and Model Sizes -- 3.3 Accuracy. 3.4 Error and Accuracy Distribution. |
Record Nr. | UNINA-9910357842303321 |
Cham : , : Springer International Publishing : , : Imprint : Springer, , 2019 | ||
![]() | ||
Lo trovi qui: Univ. Federico II | ||
|
High Performance Computing [[electronic resource] ] : 34th International Conference, ISC High Performance 2019, Frankfurt/Main, Germany, June 16–20, 2019, Proceedings / / edited by Michèle Weiland, Guido Juckeland, Carsten Trinitis, Ponnuswamy Sadayappan |
Edizione | [1st ed. 2019.] |
Pubbl/distr/stampa | Cham : , : Springer International Publishing : , : Imprint : Springer, , 2019 |
Descrizione fisica | 1 online resource (XVI, 352 p. 512 illus., 113 illus. in color.) |
Disciplina | 004.3 |
Collana | Theoretical Computer Science and General Issues |
Soggetto topico |
Software engineering
Logic design Microprocessors Computer architecture Artificial intelligence Computer networks Software Engineering Logic Design Processor Architectures Artificial Intelligence Computer Communication Networks |
ISBN | 3-030-20656-4 |
Formato | Materiale a stampa ![]() |
Livello bibliografico | Monografia |
Lingua di pubblicazione | eng |
Record Nr. | UNINA-9910337859503321 |
Cham : , : Springer International Publishing : , : Imprint : Springer, , 2019 | ||
![]() | ||
Lo trovi qui: Univ. Federico II | ||
|
High Performance Computing [[electronic resource] ] : 33rd International Conference, ISC High Performance 2018, Frankfurt, Germany, June 24-28, 2018, Proceedings / / edited by Rio Yokota, Michèle Weiland, David Keyes, Carsten Trinitis |
Edizione | [1st ed. 2018.] |
Pubbl/distr/stampa | Cham : , : Springer International Publishing : , : Imprint : Springer, , 2018 |
Descrizione fisica | 1 online resource (XV, 412 p. 177 illus.) |
Disciplina | 004.3 |
Collana | Theoretical Computer Science and General Issues |
Soggetto topico |
Electronic digital computers—Evaluation
Operating systems (Computers) Computer systems Microprocessors Computer architecture Logic design System Performance and Evaluation Operating Systems Computer System Implementation Processor Architectures Logic Design |
ISBN | 3-319-92040-5 |
Formato | Materiale a stampa ![]() |
Livello bibliografico | Monografia |
Lingua di pubblicazione | eng |
Nota di contenuto | Resource Management and Energy Efficiency -- Performance Analysis and Tools -- Exascale Networks -- Parallel Algorithms. |
Record Nr. | UNISA-996465807903316 |
Cham : , : Springer International Publishing : , : Imprint : Springer, , 2018 | ||
![]() | ||
Lo trovi qui: Univ. di Salerno | ||
|
High Performance Computing [[electronic resource] ] : 33rd International Conference, ISC High Performance 2018, Frankfurt, Germany, June 24-28, 2018, Proceedings / / edited by Rio Yokota, Michèle Weiland, David Keyes, Carsten Trinitis |
Edizione | [1st ed. 2018.] |
Pubbl/distr/stampa | Cham : , : Springer International Publishing : , : Imprint : Springer, , 2018 |
Descrizione fisica | 1 online resource (XV, 412 p. 177 illus.) |
Disciplina | 004.3 |
Collana | Theoretical Computer Science and General Issues |
Soggetto topico |
Electronic digital computers—Evaluation
Operating systems (Computers) Computer systems Microprocessors Computer architecture Logic design System Performance and Evaluation Operating Systems Computer System Implementation Processor Architectures Logic Design |
ISBN | 3-319-92040-5 |
Formato | Materiale a stampa ![]() |
Livello bibliografico | Monografia |
Lingua di pubblicazione | eng |
Nota di contenuto | Resource Management and Energy Efficiency -- Performance Analysis and Tools -- Exascale Networks -- Parallel Algorithms. |
Record Nr. | UNINA-9910349436703321 |
Cham : , : Springer International Publishing : , : Imprint : Springer, , 2018 | ||
![]() | ||
Lo trovi qui: Univ. Federico II | ||
|