检索结果-内蒙古大学图书馆

EVAdyr: A new dynamic resampling algorithm for auto-tuning noisy High Performance Computing systems

JOUrNAL OF COMPUTATIONAL SCIENCE 2025年 84卷

作者： robert-Hayek, Sophie Zertal, Soraya Couvee, Philippe Eviden BDS R&D Data Management F-38130 Echirolles France UPSaclay UVSQ Li PaRAD F-78297 Guyancourt France

Black-box auto-tuning methods have been proven to be efficient for tuning configurable computer hardware, including those encountered within the High Performance Computing (HPC) ecosystem. However, because of the shared nature of HPC clusters and the complexity of the software and hardware stacks, the measurement of the performance function can be tainted by noise during the tuning process, which can reduce and sometimes prevent the benefit of the tuning approach. A usual choice for performing the tuning in spite of these interference is to add a resampling step at each iteration to reduce uncertainty, but this approach can be timeconsuming and must be done carefully. In this paper, we propose anew resampling and filtering algorithm called EVAdyr (Efficient Value Aware dynamic resampling). Compared to the state of the art, it finds a better exploration versus exploitation trade-off by resampling only promising configuration and increases the level of confidence around the suggested solution as the tuning process advances. This algorithm was able to tune efficiently two I/O accelerators highly sensitive to interference, in two different scenarios. Compared to Standard Error dynamic resampling (SEdr), a state of the art noise reduction strategy, we show that EVAdyr is able to reduce the distance to the optimum by 93.5% and 24.7% for the two I/O accelerators respectively, as well as speed-up the experiment duration by 45.8% and 58.1% because less iterations are needed to reach the found optimum. Our results prove the importance of using noise reduction strategies whenever tuning systems running in production.

关键词： resampling Auto-tuning Noise reduction I/O accelerators HPC

来源：评论

学校读者我要写书评

暂无评论

EVAdyr: A New dynamic resampling Algorithm for Optimizing Noisy Expensive Systems 9th

EVADyR: A New Dynamic Resampling Algorithm for Optimizing No...

引用

9th International Conference on Metaheuristics and Nature Inspired Computing (META)

作者： robert-Hayek, S. Zertal, S. Couvee, P. Atos BDS R&D Data Management Echirolles France Univ UPSacaly UVSQ Guyancourt France

ISBN: (纸本)9783031692567;9783031692574

Black-box auto-tuning methods have been proven to be efficient for tuning configurable computer appliance. However, because of the shared nature and the complexity of the software and hardware stack of some systems such as cloud or HPC systems, the measurement of the performance function can be tainted by noise during the tuning process, which can reduce and sometime prevent, the benefit of auto-tuning. An usual choice is to add a resampling step at each iteration to reduce uncertainty, but this approach can be time-consuming. In this paper, we propose a new resampling and filtering algorithm called EVAdyr (Efficient Value Aware dynamic resampling). This algorithm is able to tune efficiently a prefetching strategy in the case of multiple parallel accesses. Because it finds a better exploration versus exploitation trade-off by resampling only promising parametrizations and increases the level of confidence around the suggested solution as the tuning process advances, it outperforms state of the art dynamic resampling by reducing the distance to the optimum by 93.5%, as well as speed-up the experiment duration by 45.8% because less iterations are needed to reach the found optimum. An additional proof of this study is the demonstration of the importance of using noise reduction strategies for the optimization of highly shared resources such as HPC or cloud systems.

关键词： Auto-tuning Black-box optimization High Performance Computing Stochastic Optimization resampling

来源：评论

学校读者我要写书评

暂无评论

Forecasting file lifecycles for intelligent data placement in hierarchical storage 35

Forecasting file lifecycles for intelligent data placement i...

引用

35th IEEE International Symposium on Computer Architecture and High Performance Computing (SBAC-PAd)

作者： Khelili, Adrian robert Hayek, Sophie Zertal, Soraya UPSaclay UVSQ Eviden Atos BDS R&D Data Management Li PaRAD Versailles France Eviden Atos BDS R&D Data Management Echirolles France UPSaclay UVSQ Li PaRAD Guyancourt France

ISBN: (纸本)9798350305487

The increasing disparity between computing speed and data access latency poses significant challenges in managing data storage, particularly for massively parallel supercomputers. To address this issue, storage systems have evolved into hierarchical architectures with different tiers offering varying performance, cost, and capacity based on specific hardware technologies. This heterogeneous and hierarchical nature of storage comes with the need for an optimal data placement strategy. Existing strategies have primarily approached this problem from a block perspective, focusing on analyzing application I/O behavior. However, such approaches fail to capture the contextual usage of information. To overcome this limitation, considering file-level usage patterns and adopting a file-centric perspective for data placement can leverage the context and semantics of files, leading to more efficient data placement strategies. This study proposes a novel file-based approach for data placement, focusing on the concept of file re-use and representing files through their life cycles (FLCs). The FLC of a file captures the sequence of operations it undergoes during its lifetime, referred to as FLCevents. By analyzing the time series data associated with the FLCs of actively used files, our algorithm detects repetitive patterns of FLCevents to predict future events using pattern matching and facilitate the anticipation of file movement. This allows for optimal data placement that aligns with the expected near-future usage. To validate our approach, we conducted experiments using traces extracted from representative high-performance computing (HPC) applications, namely NEMO, NAMd, LQCd, and an IO-Benchmark. The results we obtained are highly promising, demonstrating that in term of file prediction accuracy our proposed F-LrU (File-based Least recently Used) achieves from 77% to 55% precision in most difficult scenarios to 100% for the others. Also, it is at least as effective as tradit

关键词： I/O Prediction File lifecycles Time Series data temperature Hierarchical storage High Performance Computing

来源：评论

学校读者我要写书评

暂无评论

drug use investigation on the safety and efficacy of tigecycline in Japan (all-case post-marketing surveillance)

引用

JOUrNAL OF INFECTION ANd CHEMOTHErAPY 2022年第7期28卷 866-874页

作者： Ohashi, Takahisa Sugiyama, Noriko Watanabe, Toshiya Uryu, Taku Yoshinaga, Yukari Pfizer Japan Inc Med Affairs Dept Hosp Business Unit Tokyo Japan Pfizer R&D Biometr & Data Management Dept Clin Stat Grp 2 Tokyo Japan Pfizer R&D Data Monitoring & Management Grp Biometr & Data Management Dept Tokyo Japan Pfizer R&D Stat Programming & Anal Grp Biometr & Data Management Dept Tokyo Japan Pfizer R&D Postmkt Study Strategy & Management Dept Postmkt Study Strategy & Management Grp 1 Tokyo Japan

Introduction: We conducted a drug use investigation to investigate the safety and efficacy of tigecycline, which has been approved for clinical use for the treatment of multidrug-resistant gram-negative infections in Japan. Methods: This was an open-label, observational, multicenter cohort study that included all patients who received tigecycline. results: A total of 116 patients were registered between december 2012 and April 2016 and all of them were evaluated for safety and efficacy. Among them, 64 patients aged >= 65 years (55.2%) and five children aged < 15 years (4.3%) were included. Of these patients, 47 (40.5%) met the approved indications of tigecycline. Adverse drug reactions (Adrs) were observed in 41 patients (35.3%) with a total of 74 events. Serious Adrs were observed in 15 patients (12.93%) with a total of 33 events. There were 42 deaths, and 6 of these were considered to be caused by Adrs. Among the 116 patients, 65 achieved clinical response at the end of the observation period, and the efficacy rate was 73.9%. Furthermore, 46 patients were assessed as "cure " at the test of cure visit, and the cure rate was 59.0%. The eradication rate was 47.5% at the end of the observation period. Classified by pathogenic bacteria, the eradication rate of patients infected with the approved pathogens was 54.5%. Conclusions: Tigecycline was well-tolerated, and no additional safety concerns were noted. It was effective considering that most patients had poor physical conditions. The overall benefit-risk balance of tigecycline was favorable.

关键词： Tigecycline Post-marketing surveillance Multidrug-resistant bacteria Efficacy Safety Adverse reaction

来源：评论

学校读者我要写书评

暂无评论

Pattern Matching to improve Accuracy and Efficiency of File Lifecycles Forecasting 14

Pattern Matching to improve Accuracy and Efficiency of File ...

引用

14th International Conference on Intelligent Systems: Theories and Applications, SITA 2023

作者： Khelili, Adrian Hayek, Sophie robert Zertal, Soraya UPSaclay-UVSQ Eviden Atos R&d Data Management Li-PaRAD France Eviden Atos R&d Data Management Echirolles France UPSaclay-UVSQ Li-PaRAD Guyancourt France

ISBN: (纸本)9798350308211

Time series prediction is a crucial task with applications in various domains. In this research, we introduce a transfer learning approach that goes beyond the historical time series of individual instances to enhance prediction performance. Traditional time series prediction methods often focus solely on historical data from a single time series, limiting their ability to capture already known patterns and behaviors. Our proposed method addresses this limitation by constructing a curated database of time series representative of the behavior to predict. We apply our suggested method to the prediction of the I/O (Input/Output) behavior of scientific applications such as NEMO, NAMd, and LQCd. Our non-redundant database contains diverse time series, allowing us to extract valuable insights and patterns that might not be discernible from individual series. By leveraging similarities and patterns across those time series, our approach improves prediction accuracy and robustness. The results we obtained are highly promising, demonstrating that in term of time serie ranking prediction accuracy our proposed inter-timeseries method achieves from 72.9% to 83.8% accuracy in most difficult scenarios to 97% for the others. To compare our solution to existing methods, we show an improvement of 10.1% accuracy compared to ArIMA. © 2023 IEEE.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Prediction of cardiovascular risk factors from retinal fundus photographs: Validation of a deep learning algorithm in a prospective non-interventional study in Kenya

引用

dIABETES OBESITY & METABOLISM 2024年第7期26卷 2722-2731页

作者： White, Tom Selvarajah, Viknesh Wolfhagen-Sand, Fredrik Svangard, Nils Mohankumar, Gayathri Fenici, Peter rough, Kathryn Onyango, Nelson Lyons, Kendall Mack, Christina Nduba, Videlis Saleh, Mansoor Noorali Abayo, Innocent Siddiqui, Afrah Majdanska-Strzalka, Malgorzata Kaszubska, Katarzyna Hegelund-Myrback, Tove Esterline, russell Manzur, Antonio Parker, Victoria E. r. AstraZeneca Data Sci & Adv Analyt Data Sci Artificial Intelligence R&D Cambridge England AstraZeneca Res & Early Dev Cardiovasc Renal & Metab CVRM BioPharmaceut R&D Cambridge England AstraZeneca R&D Data Sci & Adv Analyt Data Sci & Artificial Intelligence Gothenburg Sweden Ctr Artificial Intelligence Data Sci & Artificial Intelligence R&D AstraZeneca Gaithersburg MD USA Catholic Univ Sch Med & Surg Rome Italy Biomagnetism & Clin Physiol Int Ctr BACP Rome Italy AstraZeneca Med Affairs BioPharmaceut Milan Italy IQVIA Durham NC USA Kenya Govt Med Res Ctr Nairobi Kenya Aga Khan Univ Hosp Clin Res Unit Nairobi Kenya AstraZeneca BioPharmaceut Med Cambridge England CVRM Clin Operat Biopharmaceut R&D AstraZeneca Warsaw Poland Global Portfolio & Project Management Early CVRM&NS R&D AstraZeneca Gothenburg Sweden AstraZeneca Cardiovasc Renal & Metab CVRM BioPharmaceut R&D Gaithersburg MD USA

AimHypertension and diabetes mellitus (dM) are major causes of morbidity and mortality, with growing burdens in low-income countries where they are underdiagnosed and undertreated. Advances in machine learning may provide opportunities to enhance diagnostics in settings with limited medical *** and MethodsA non-interventional study was conducted to develop and validate a machine learning algorithm to estimate cardiovascular clinical and laboratory parameters. At two sites in Kenya, digital retinal fundus photographs were collected alongside blood pressure (BP), laboratory measures and medical history. The performance of machine learning models, originally trained using data from the UK Biobank, were evaluated for their ability to estimate BP, glycated haemoglobin, estimated glomerular filtration rate and diagnoses from fundus *** total, 301 participants were enrolled. Compared with the UK Biobank population used for algorithm development, participants from Kenya were younger and would probably report Black/African ethnicity, with a higher body mass index and prevalence of dM and hypertension. The mean absolute error was comparable or slightly greater for systolic BP, diastolic BP, glycated haemoglobin and estimated glomerular filtration rate. The model trained to identify dM had an area under the receiver operating curve of 0.762 (0.818 in the UK Biobank) and the hypertension model had an area under the receiver operating curve of 0.765 (0.738 in the UK Biobank).ConclusionsIn a Kenyan population, machine learning models estimated cardiovascular parameters with comparable or slightly lower accuracy than in the population where they were trained, suggesting model recalibration may be appropriate. This study represents an incremental step toward leveraging machine learning to make early cardiovascular screening more accessible, particularly in resource-limited settings.

关键词： cardiovascular screening diabetes mellitus hypertension machine learning predictive modelling retinal fundus photographs UK Biobank

来源：评论

学校读者我要写书评

暂无评论

Group buying with consumer disappointment at failed deals

引用

TrANSPOrTATION rESEArCH PArT E-LOGISTICS ANd TrANSPOrTATION rEVIEW 2025年 194卷

作者： Wang, Jie Ma, Benedict Jun Yang, Yanyi Cheng, Chun-Hung Kuo, Yong-Hong Southeast Univ Sch Econ & Management Nanjing Peoples R China Univ Hong Kong Dept Data & Syst Engn Hong Kong Peoples R China Chinese Univ Hong Kong Dept Syst Engn & Engn Management Hong Kong Peoples R China Logist & Supply Chain MultiTech R&D Ctr Ltd Hong Kong Peoples R China

In group buying (GB), the retailer launches a deal with a discounted product price and a minimum group size requirement. Strategic consumers then determine whether to sign up for the GB deal or purchase the product at the regular price immediately. If GB fails, disappointed GB participants perceive a negative psychological utility and decide whether or not to buy it again at the regular price. Considering the disappointment caused by a GB failure, in our basic model, we formulate a two-period game to study the retailer's optimal pricing decisions and consumers' purchasing strategies. By deriving the likelihood function of a consumer signing up for the GB deal and utilizing rational expectations theory, we characterize how consumers form their own beliefs about the GB success rate. We find that consumer sentiment toward a failed GB deal serves an important role in a GB deal. Specifically, there exists a threshold of the disappointment factor where the retailer's profit and consumers' purchase decisions may change, and consumer leakage and retention may occur. We prove the existence of the subgame perfect Nash equilibrium and outline how the retailer should design the GB schedule. Our study shows that if properly designed, GB is an effective strategy to enhance consumer interests and improve profit. Moreover, a big discount should be offered when the disappointment factor is significant. When the consumer sentiment toward a failed GB deal is insignificant, the retailer should launch a GB deal;otherwise, he should provide the regular sales channel only. We conduct numerical experiments to study the impacts of different factors in a GB deal. Our key results continue to hold in several extensions from our basic model: retailer competition, observable secured group size, and social interactions between consumers.

关键词： Group buying Strategic consumer behavior disappointment factor

来源：评论

学校读者我要写书评

暂无评论

Use of pharmacodynamic modeling for Bayesian information borrowing in pediatric clinical trials

引用

JOUrNAL OF BIOPHArMACEUTICAL STATISTICS 2023年第6期33卷 726-736页

作者： Sebastien, Bernard Cheung, S. Y. Amy Corriol-rohou, Solange Gamalo-Siebers, Margaret Jreich, rana Krishna, rajesh Liu, Jing Sanofi Data & Data Sci R&D Chilly Mazarin France Certara USA Inc Integrated Drug Dev Princeton NJ USA AstraZeneca Regulatory Affairs & Policy Global Regulatory Policy R&D Paris France Pfizer Global Biometr & Data Management Innovat Hlth Collegeville PA USA Certara USA Inc Clin Pharmacol Princeton NJ USA Pfizer Inc Clin Pharmacol Groton CT USA Sanofi Data & Data Sci R&D 1 Ave Pierre Brossolette F-91385 Chilly Mazarin France

The use of Bayesian methodology to design and analyze pediatric efficacy trials is one of the possible options to reduce their sample size. This reduction of the sample size results from the use of an informative prior for the parameters of interest. In most of the applications, the principle of 'information borrowing' from adults' trials is applied, which means that the informative prior is constructed using efficacy results in adult of the drug under investigation. This implicitly assumes similarity in efficacy between the selected pediatric dose and the efficacious dose in adults. The goal of this article is to propose a method to construct prior distribution for the parameter of interest, not directly constructed from the efficacy results of the efficacious dose in adult patients but using pharmacodynamic modeling of a bridging biomarker using early phase pediatric data. When combined with a model bridging the biomarker with the clinical endpoints, the prior is constructed using a variational method after simulation of the parameters of interest. A use case application illustrates how the method can be used to construct a realistic informative prior.

关键词： Bayesian prior pediatric studies PKPd modeling simulations variational method

来源：评论

学校读者我要写书评

暂无评论

GrIOt: Graph-based Modeling of HPC Application I/O Call Stacks for Predictive Prefetch

GrIOt: Graph-based Modeling of HPC Application I/O Call Stac...

引用

2023 International Conference on High Performance Computing, Network, Storage, and Analysis, SC Workshops 2023

作者： Nicolas, Louis-Marie Mimouni, Salim Couvée, Philippe Boukhobza, Jalil Ensta Bretagne Lab-STICC Cnrs Umr 6285 Atos Bds R&d Data Management Grenoble France Atos Bds R&d Data Management Grenoble France Ensta Bretagne Lab-STICC Cnrs Umr 6285 Brest France

ISBN: (纸本)9798400707858

Modern High Performance Computing (HPC) storage systems use heterogeneous storage technologies organized in tiers to find a compromise between capacity, performance, and cost. In these systems, prefetching is a common technique used to move the right data at the right moment from a slow to a fast tier to improve the overall performance while using the costly high-performance tier only when needed. Effective prefetching requires precise knowledge of the application I/O patterns. This knowledge can be extracted through the source code, I/O tracing tools or I/O functions call stacks. State-of-the-art solutions based on the latter approach mainly focus on applications with regular I/O profiles to avoid scalability issues due to the grammar-based techniques used. In this paper, we present an approach based on I/O call stacks that models I/O patterns for both regular and irregular applications, thanks to the use of directed graphs. We present different models for prefetching. Our models were used to predict the next I/O call stack on two real HPC applications and one synthetic workload with an accuracy of up to 98%, while keeping a low overheadt. © 2023 ACM.

关键词： directed graphs

来源：评论

学校读者我要写书评

暂无评论

Simultaneous inhibition of dNA-PK and Polϴ improves integration efficiency and precision of genome editing

引用

NATUrE COMMUNICATIONS 2023年第1期14卷 1-18页

作者： Wimberger, Sandra Akrap, Nina Firth, Mike Brengdahl, Johan Engberg, Susanna Schwinn, Marie K. Slater, Michael r. Lundin, Anders Hsieh, Pei-Pei Li, Songyuan Cerboni, Silvia Sumner, Jonathan Bestas, Burcu Schiffthaler, Bastian Magnusson, Bjoern di Castro, Silvio Iyer, Preeti Bohlooly-Y, Mohammad Machleidt, Thomas rees, Steve Engkvist, Ola Norris, Tyrell Cadogan, Elaine B. Forment, Josep V. Svikovic, Sasa Akcakaya, Pinar Taheri-Ghahfarokhi, Amir Maresca, Marcello AstraZeneca BioPharmaceut R&D Genome Engn Discovery Sci Gothenburg Sweden Univ Gothenburg Dept Chem & Mol Biol Gothenburg Sweden AstraZeneca BioPharmaceut R&D Discovery Sci Data Sci & Quantitat Biol Cambridge England AstraZeneca BioPharmaceut R&D Cell Assay Dev Discovery Sci Gothenburg Sweden AstraZeneca BioPharmaceut R&D Cell Engn Sweden Discovery Sci Gothenburg Sweden Promega Corp Madison WI USA AstraZeneca BioPharmaceut R&D Translat Genom Discovery Sci Gothenburg Sweden AstraZeneca BioPharmaceut R&D Translat Sci & Expt Med Res & Early DevResp & Immunol R&I Gothenburg Sweden AstraZeneca BioPharmaceut R&D Cell Immunol Discovery Sci Cambridge England AstraZeneca BioPharmaceut R&D Discovery Sci Data Sci & Quantitat Biol Gothenburg Sweden AstraZeneca BioPharmaceut R&D Compound Synth & Management Discovery Sci Gothenburg Sweden AstraZeneca BioPharmaceut R&D Mol AI Discovery Sci Gothenburg Sweden AstraZeneca Discovery Sci BioPharmaceut R&D Cambridge England AstraZeneca Biosci Early Oncol Cambridge England

Genome editing, specifically CrISPr/Cas9 technology, has revolutionized biomedical research and offers potential cures for genetic diseases. despite rapid progress, low efficiency of targeted dNA integration and generation of unintended mutations represent major limitations for genome editing applications caused by the interplay with dNA double-strand break repair pathways. To address this, we conduct a large-scale compound library screen to identify targets for enhancing targeted genome insertions. Our study reveals dNA-dependent protein kinase (dNA-PK) as the most effective target to improve CrISPr/Cas9-mediated insertions, confirming previous findings. We extensively characterize AZd7648, a selective dNA-PK inhibitor, and find it to significantly enhance precise gene editing. We further improve integration efficiency and precision by inhibiting dNA polymerase theta (Pol ϴ). The combined treatment, named 2iHdr, boosts templated insertions to 80% efficiency with minimal unintended insertions and deletions. Notably, 2iHdr also reduces off-target effects of Cas9, greatly enhancing the fidelity and performance of CrISPr/Cas9 gene editing. Low efficiency of target dNA integration remains a challenge in genome engineering. Here the authors perform large-scale compound library and genetic screens to identify targets that enhance gene editing: they see that combined dNA-PK and Pol ϴ inhibition with potent compounds increases editing efficiency and precision.

关键词： CrISPr-Cas9 genome editing Homologous recombination

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：