scholarly journals Concurrency and Interference Analysis of Kernels on GPUs

2021 ◽  
Author(s):  
Pablo Carvalho ◽  
Lúcia Maria De A. Drummond ◽  
Cristiana Bentes
Keyword(s):  
Large Scale ◽  
Data Centers ◽  
Heterogeneous Systems ◽  
System Throughput ◽  
Concurrent Execution ◽  
Large Scale Data ◽  
Resource Requirements ◽  
Cloud Environments ◽  
The Impact ◽  
Scale Data

Heterogeneous systems employing CPUs and GPUs are becoming increasingly popular in large-scale data centers and cloud environments. In these platforms, sharing a GPU across different applications is an important feature to improve hardware utilization and system throughput. However, under scenarios where GPUs are competitively shared, some challenges arise. The decision on the simultaneous execution of different kernels is made by the hardware and depends on the kernels resource requirements. Besides that, it is very difficult to understand all the hardware variables involved in the simultaneous execution decisions, in order to describe a formal allocation method. In this work, we studied the impact that kernel resource requirements have in concurrent execution and used machine learning (ML) techniques to infer the interference caused by the concurrent execution, and to classify the slowdown that results from this interference. The ML techniques were analyzed over the GPU benchmark suites, Rodinia, Parboil and SHOC. Our results showed that, from the features selected in the analysis, the number of blocks per grid, number of threads per block, and number of registers are the resource consumption features that most affect the performance of the concurrent execution.

scholarly journals Multi-GPU approach to global induction of classification trees for large-scale data mining

2021 ◽  
Author(s):  
Krzysztof Jurczuk ◽  
Marcin Czajkowski ◽  
Marek Kretowski
Keyword(s):  
Data Mining ◽  
Large Scale ◽  
Real Life ◽  
Population Based ◽  
Tree Structure ◽  
Global Approach ◽  
Data Parallel ◽  
Large Scale Data ◽  
The Impact ◽  
Scale Data

AbstractThis paper concerns the evolutionary induction of decision trees (DT) for large-scale data. Such a global approach is one of the alternatives to the top-down inducers. It searches for the tree structure and tests simultaneously and thus gives improvements in the prediction and size of resulting classifiers in many situations. However, it is the population-based and iterative approach that can be too computationally demanding to apply for big data mining directly. The paper demonstrates that this barrier can be overcome by smart distributed/parallel processing. Moreover, we ask the question whether the global approach can truly compete with the greedy systems for large-scale data. For this purpose, we propose a novel multi-GPU approach. It incorporates the knowledge of global DT induction and evolutionary algorithm parallelization together with efficient utilization of memory and computing GPU’s resources. The searches for the tree structure and tests are performed simultaneously on a CPU, while the fitness calculations are delegated to GPUs. Data-parallel decomposition strategy and CUDA framework are applied. Experimental validation is performed on both artificial and real-life datasets. In both cases, the obtained acceleration is very satisfactory. The solution is able to process even billions of instances in a few hours on a single workstation equipped with 4 GPUs. The impact of data characteristics (size and dimension) on convergence and speedup of the evolutionary search is also shown. When the number of GPUs grows, nearly linear scalability is observed what suggests that data size boundaries for evolutionary DT mining are fading.

On Workload-Aware DRAM Failure Prediction in Large-Scale Data Centers

2021 ◽  
Author(s):  
Xingyi Wang ◽  
Yu Li ◽  
Yiquan Chen ◽  
Shiwen Wang ◽  
Yin Du ◽  
...  
Keyword(s):  
Large Scale ◽  
Data Centers ◽  
Failure Prediction ◽  
Large Scale Data ◽  
Scale Data

DRAM Failure Prediction in Large-Scale Data Centers

2021 ◽  
Author(s):  
Fengyuan Yu ◽  
Hongzuo Xu ◽  
Songlei Jian ◽  
Chenlin Huang ◽  
Yijie Wang ◽  
...  
Keyword(s):  
Large Scale ◽  
Data Centers ◽  
Failure Prediction ◽  
Large Scale Data ◽  
Scale Data

Data Warehousing Requirements Collection and Definition

2012 ◽  
pp. 261-271
Author(s):  
Nenad Jukic ◽  
Miguel Velasco
Keyword(s):  
Data Warehouse ◽  
Large Scale ◽  
Data Warehousing ◽  
System Development ◽  
Large Scale Data ◽  
Typical Data ◽  
Potential Risks ◽  
Real Scenario ◽  
The Impact ◽  
Scale Data

Defining data warehouse requirements is widely recognized as one of the most important steps in the larger data warehouse system development process. This paper examines the potential risks and pitfalls within the data warehouse requirement collection and definition process. A real scenario of a large-scale data warehouse implementation is given, and details of this project, which ultimately failed due to inadequate requirement collection and definition process, are described. The presented case underscores and illustrates the impact of the requirement collection and definition process on the data warehouse implementation, while the case is analyzed within the context of the existing approaches, methodologies, and best practices for prevention and avoidance of typical data warehouse requirement errors and oversights.

Data Warehousing Requirements Collection and Definition

2010 ◽  
Vol 1 (3) ◽  
pp. 66-76
Author(s):  
Nenad Jukic ◽  
Miguel Velasco
Keyword(s):  
Data Warehouse ◽  
Large Scale ◽  
Development Process ◽  
System Development ◽  
Large Scale Data ◽  
Typical Data ◽  
Potential Risks ◽  
Real Scenario ◽  
The Impact ◽  
Scale Data

Defining data warehouse requirements is widely recognized as one of the most important steps in the larger data warehouse system development process. This paper examines the potential risks and pitfalls within the data warehouse requirement collection and definition process. A real scenario of a large-scale data warehouse implementation is given, and details of this project, which ultimately failed due to inadequate requirement collection and definition process, are described. The presented case underscores and illustrates the impact of the requirement collection and definition process on the data warehouse implementation, while the case is analyzed within the context of the existing approaches, methodologies, and best practices for prevention and avoidance of typical data warehouse requirement errors and oversights.

Multi-omics investigations within the Phylum Mollusca, Class Gastropoda: from ecological application to breakthrough phylogenomic studies

2019 ◽  
Author(s):  
Anne H Klein ◽  
Kaylene R Ballard ◽  
Kenneth B Storey ◽  
Cherie A Motti ◽  
Min Zhao ◽  
...  
Keyword(s):  
Large Scale ◽  
Environmental Stressors ◽  
Data Organization ◽  
Large Scale Data ◽  
Huge Data ◽  
The Future ◽  
Ecological Application ◽  
The Impact ◽  
Generation Sequencing ◽  
Scale Data

Abstract Gastropods are the largest and most diverse class of mollusc and include species that are well studied within the areas of taxonomy, aquaculture, biomineralization, ecology, microbiome and health. Gastropod research has been expanding since the mid-2000s, largely due to large-scale data integration from next-generation sequencing and mass spectrometry in which transcripts, proteins and metabolites can be readily explored systematically. Correspondingly, the huge data added a great deal of complexity for data organization, visualization and interpretation. Here, we reviewed the recent advances involving gastropod omics (‘gastropodomics’) research from hundreds of publications and online genomics databases. By summarizing the current publicly available data, we present an insight for the design of useful data integrating tools and strategies for comparative omics studies in the future. Additionally, we discuss the future of omics applications in aquaculture, natural pharmaceutical biodiscovery and pest management, as well as to monitor the impact of environmental stressors.

Towards a New Data Replication Management in Cloud Systems

2019 ◽  
Vol 10 (2) ◽  
pp. 1-20
Author(s):  
Abdenour Lazeb ◽  
Riad Mokadem ◽  
Ghalem Belalem
Keyword(s):  
Large Scale ◽  
Data Replication ◽  
Cloud Provider ◽  
Large Scale Data ◽  
Dynamic Replication ◽  
Replication Strategy ◽  
Multiple Copies ◽  
Cloud Environments ◽  
Real Challenge ◽  
Scale Data

Applications produce huge volumes of data that are distributed on remote and heterogeneous sites. This generates problems related to access and sharing data. As a result, managing data in large-scale environments is a real challenge. In this context, large-scale data management systems often use data replication, a well-known technique that treats generated problems by storing multiple copies of data, called replicas, across multiple nodes. Most of the replication strategies in these environments are difficult to adapt to cloud environments. They aim to achieve the best performance of the system without meeting the important objectives of the cloud provider. This article proposes a new dynamic replication strategy. The proposed algorithm significantly improves provider gain without neglecting customer satisfaction.

Sign in / Sign up

Export Citation Format

Share Document