首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 15 毫秒
1.
MPI collective communication operations to distribute or gather data are used for many parallel applications from scientific computing, but they may lead to scalability problems since their execution times increase with the number of participating processors. In this article, we show how the execution time of collective communication operations can be improved significantly by an internal restructuring based on orthogonal processor structures with two or more levels. The execution time of operations like MPI_Bcast() or MPI_Allgather() can be reduced by 40% and 70% on a dual Xeon cluster and a Beowulf cluster with single-processor nodes. But also on a Cray T3E a significant performance improvement can be obtained by a careful selection of the processor structure. The use of these optimized communication operations can reduce the execution time of data parallel implementations of complex application programs significantly without requiring any other change of the computation and communication structure. We present runtime functions for the modeling of two-phase realizations and verify that these runtime functions can predict the execution time both for communication operations in isolation and in the context of application programs.  相似文献   

2.
The pay-as-you-go pricing model and the illusion of unlimited resources in the Cloud initiate the idea to provision services elastically. Elastic provisioning of services allocates/de-allocates resources dynamically in response to the changes of the workload. It minimizes the service provisioning cost while maintaining the desired service level objectives (SLOs). Model-predictive control is often used in building such elasticity controllers that dynamically provision resources. However, they need to be trained, either online or offline, before making accurate scaling decisions. The training process involves tedious and significant amount of work as well as some expertise, especially when the model has many dimensions and the training granularity is fine, which is proved to be essential in order to build an accurate elasticity controller. In this paper, we present OnlineElastMan, which is a self-trained proactive elasticity manager for cloud-based storage services. It automatically evolves itself while serving the workload. Experiments using OnlineElastMan with Cassandra indicate that OnlineElastMan continuously improves its provision accuracy, i.e., minimizing provisioning cost and SLO violations, under various workload patterns.  相似文献   

3.
Cheng  Feng  Huang  Yifeng  Tanpure  Bhavana  Sawalani  Pawan  Cheng  Long  Liu  Cong 《Cluster computing》2022,25(1):619-631

As the services provided by cloud vendors are providing better performance, achieving auto-scaling, load-balancing, and optimized performance along with low infrastructure maintenance, more and more companies migrate their services to the cloud. Since the cloud workload is dynamic and complex, scheduling the jobs submitted by users in an effective way is proving to be a challenging task. Although a lot of advanced job scheduling approaches have been proposed in the past years, almost all of them are designed to handle batch jobs rather than real-time workloads, such as that user requests are submitted at any time with any amount of numbers. In this work, we have proposed a Deep Reinforcement Learning (DRL) based job scheduler that dispatches the jobs in real time to tackle this problem. Specifically, we focus on scheduling user requests in such a way as to provide the quality of service (QoS) to the end-user along with a significant reduction of the cost spent on the execution of jobs on the virtual instances. We have implemented our method by Deep Q-learning Network (DQN) model, and our experimental results demonstrate that our approach can significantly outperform the commonly used real-time scheduling algorithms.

  相似文献   

4.
In this paper, an autonomic performance management approach is introduced that can be applied to a general class of web services deployed in large scale distributed environment. The proposed approach utilizes traditional large scale control-based algorithms by using interaction balance approach in web service environment for managing the response time and the system level power consumption. This approach is developed in a generic fashion that makes it suitable for web service deployments, where web service performance can be adjusted by using a finite set of control inputs. This approach maintains the service level agreements, maximizes the revenue, and minimizes the infrastructure operating cost. Additionally, the proposed approach is fault-tolerant with respect to the failures of the computing nodes inside the distributed deployment. Moreover, the computational overhead of the proposed approach can also be managed by using appropriate value of configuration parameters during its deployment.  相似文献   

5.
Live migration of virtual machine (VM) provides a significant benefit for virtual server mobility without disrupting service. It is widely used for system management in virtualized data centers. However, migration costs may vary significantly for different workloads due to the variety of VM configurations and workload characteristics. To take into account the migration overhead in migration decision-making, we investigate design methodologies to quantitatively predict the migration performance and energy consumption. We thoroughly analyze the key parameters that affect the migration cost from theory to practice. We construct application-oblivious models for the cost prediction by using learned knowledge about the workloads at the hypervisor (also called VMM) level. This should be the first kind of work to estimate VM live migration cost in terms of both performance and energy in a quantitative approach. We evaluate the models using five representative workloads on a Xen virtualized environment. Experimental results show that the refined model yields higher than 90% prediction accuracy in comparison with measured cost. Model-guided decisions can significantly reduce the migration cost by more than 72.9% at an energy saving of 73.6%.  相似文献   

6.
The complexity and requirements of web applications are increasing in order to meet more sophisticated business models (web services and cloud computing, for instance). For this reason, characteristics such as performance, scalability and security are addressed in web server cluster design. Due to the rising energy costs and also to environmental concerns, energy consumption in this type of system has become a main issue. This paper shows energy consumption reduction techniques that use a load forecasting method, combined with DVFS (Dynamic Voltage and Frequency Scaling) and dynamic configuration techniques (turning servers on and off), in a soft real-time web server clustered environment. Our system promotes energy consumption reduction while maintaining user’s satisfaction with respect to request deadlines being met. The results obtained show that prediction capabilities increase the QoS (Quality of Service) of the system, while maintaining or improving the energy savings over state-of-the-art power management mechanisms. To validate this predictive policy, a web application running a real workload profile was deployed in an Apache server cluster testbed running Linux.  相似文献   

7.

Over the last decades, web services are used for performing specific tasks demanded by users. The most important task of service’s classification system is to match an anonymous input service with the stored pre-classified web services. The most challenging issue is that web services are currently organized and classified according to syntax while the context of the requested service is ignored. Due to this motivation, Cloud-based Classification Methodology is proposed as it presents a new methodology based on semantic web service’s classification. Furthermore, cloud computing is used for not only storing but also allocating the high scale of web services with both high availability and accessibility. Fog technology is employed to reduce the latency and to speed up response time. The experimental results using the suggested methodology show a better performance of the proposed system regarding both precision and accuracy in comparison with most of the methods discussed in the literature of the current study.

  相似文献   

8.
This paper proposes a system named AWSCS (Automatic Web Service Composition System) to evaluate different approaches for automatic composition of Web services, based on QoS parameters that are measured at execution time. The AWSCS is a system to implement different approaches for automatic composition of Web services and also to execute the resulting flows from these approaches. Aiming at demonstrating the results of this paper, a scenario was developed, where empirical flows were built to demonstrate the operation of AWSCS, since algorithms for automatic composition are not readily available to test. The results allow us to study the behaviour of running composite Web services, when flows with the same functionality but different problem-solving strategies were compared. Furthermore, we observed that the influence of the load applied on the running system as the type of load submitted to the system is an important factor to define which approach for the Web service composition can achieve the best performance in production.  相似文献   

9.
Cloud computing is becoming the new generation computing infrastructure, and many cloud vendors provide different types of cloud services. How to choose the best cloud services for specific applications is very challenging. Addressing this challenge requires balancing multiple factors, such as business demands, technologies, policies and preferences in addition to the computing requirements. This paper recommends a mechanism for selecting the best public cloud service at the levels of Infrastructure as a Service (IaaS) and Platform as a Service (PaaS). A systematic framework and associated workflow include cloud service filtration, solution generation, evaluation, and selection of public cloud services. Specifically, we propose the following: a hierarchical information model for integrating heterogeneous cloud information from different providers and a corresponding cloud information collecting mechanism; a cloud service classification model for categorizing and filtering cloud services and an application requirement schema for providing rules for creating application-specific configuration solutions; and a preference-aware solution evaluation mode for evaluating and recommending solutions according to the preferences of application providers. To test the proposed framework and methodologies, a cloud service advisory tool prototype was developed after which relevant experiments were conducted. The results show that the proposed system collects/updates/records the cloud information from multiple mainstream public cloud services in real-time, generates feasible cloud configuration solutions according to user specifications and acceptable cost predication, assesses solutions from multiple aspects (e.g., computing capability, potential cost and Service Level Agreement, SLA) and offers rational recommendations based on user preferences and practical cloud provisioning; and visually presents and compares solutions through an interactive web Graphical User Interface (GUI).  相似文献   

10.
Why are marine species where they are? The scientific community is faced with an urgent need to understand aquatic ecosystem dynamics in the context of global change. This requires development of scientific tools with the capability to predict how biodiversity, natural resources, and ecosystem services will change in response to stressors such as climate change and further expansion of fishing. Species distribution models and ecosystem models are two methodologies that are being developed to further this understanding. To date, these methodologies offer limited capabilities to work jointly to produce integrated assessments that take both food web dynamics and spatial-temporal environmental variability into account. We here present a new habitat capacity model as an implementation of the spatial-temporal model Ecospace of the Ecopath with Ecosim approach. The new model offers the ability to drive foraging capacity of species from the cumulative impacts of multiple physical, oceanographic, and environmental factors such as depth, bottom type, temperature, salinity, oxygen concentrations, and so on. We use a simulation modeling procedure to evaluate sampling characteristics of the new habitat capacity model. This development bridges the gap between envelope environmental models and classic ecosystem food web models, progressing toward the ability to predict changes in marine ecosystems under scenarios of global change and explicitly taking food web direct and indirect interactions into account.  相似文献   

11.
With the development of human–machine systems, there has been a growing concern about the consequences of operator performance breakdown under excessive level of workload, especially in safety-critical situations. Assessment and detection of the operator functional state (OFS) enable us to predict the high operational risks of operator. This paper adopts the psychophysiological signals and task performance measures to evaluate OFS under different levels of mental workload. Four indices extracted from electrocardiogram and electroencephalogram, including heart rate (HR), ratio of the standard deviation to the average of HR segment, task load indices (TLI1 and TLI2), are chosen as the inputs of the proposed model. A technique of differential evolution with ant colony search (DEACS) is developed to optimize the parameters of Adaptive-Network-based Fuzzy Inference System (ANFIS). The optimized ANFIS model is employed to estimate the OFS under a series of process control tasks on a simulated software platform of AUTOmation-enhanced Cabin Air Management System. The results showed that the proposed adaptive fuzzy model based on ANFIS and DEACS algorithm is applicable for the operator functional state assessment.  相似文献   

12.
Due to the increase of the diversity of parallel architectures, and the increasing development time for parallel applications, performance portability has become one of the major considerations when designing the next generation of parallel program execution models, APIs, and runtime system software. This paper analyzes both code portability and performance portability of parallel programs for fine-grained multi-threaded execution and architecture models. We concentrate on one particular event-driven fine-grained multi-threaded execution model—EARTH, and discuss several design considerations of the EARTH model and runtime system that contribute to the performance portability of parallel applications. We believe that these are important issues for future high end computing system software design. Four representative benchmarks were conducted on several different parallel architectures, including two clusters listed in the 23rd supercomputer TOP500 list. The results demonstrate that EARTH based programs can achieve robust performance portability across the selected hardware platforms without any code modification or tuning.  相似文献   

13.
Economic escape models predict escape decisions of prey which are approached by predators. Flight initiation distance (FID, predator–prey distance when prey begins to flee) and distance fled (DF) are major variables used to characterize escape responses. In optimal escape theory, FID increases as cost of not fleeing also increases. Moreover, FID decreases as cost of fleeing increases, due to lost opportunities to perform activities that may increase fitness. Finally, FID further increases as the prey's fitness increases. Some factors, including parasitism, may affect more than one of these predictors of FID. Initially, parasitized prey may have lower fitness as well as impaired locomotor ability, which would avoid predation and/or reduce their foraging ability, further decreasing the opportunity of fleeing. For example, if parasites decrease body condition, prey fitness is reduced and escape ability may be impaired. Hence, the overall influence of parasitism on FID is difficult to predict. We examined relationships between escape decisions and different traits: parasite load, body size and body condition in the Balearic lizard, Podarcis lilfordi. Lizards that showed higher haemogregarines load had longer FID and shorter DF. Although results did not confirm our initial predictions made on the basis of optimal escape theory, our findings suggest that parasites can alter several aspects of escape behaviour in a complex way.  相似文献   

14.
Cloud computing is a computational model in which resource providers can offer on-demand services to clients in a transparent way. However, to be able to guarantee quality of service without limiting the number of accepted requests, providers must be able to dynamically manage the available resources so that they can be optimized. This dynamic resource management is not a trivial task, since it involves meeting several challenges related to workload modeling, virtualization, performance modeling, deployment and monitoring of applications on virtualized resources. This paper carries out a performance evaluation of a module for resource management in a cloud environment that includes handling available resources during execution time and ensuring the quality of service defined in the service level agreement. An analysis was conducted of different resource configurations to define which dimension of resource scaling has a real influence on client requests. The results were used to model and implement a simulated cloud system, in which the allocated resource can be changed on-the-fly, with a corresponding change in price. In this way, the proposed module seeks to satisfy both the client by ensuring quality of service, and the provider by ensuring the best use of resources at a fair price.  相似文献   

15.
Threats to ecosystems globally from anthropogenic disturbance and climate change requires us to urgently identify the most sensitive biological communities to ensure they are effectively preserved. It is for this reason that understanding and predicting food web stability has been topical within ecology. Food web stability is a multi-faceted concept that represents the ability of a food web to maintain its integrity following disturbance, it includes resistance, resilience and fragility. In this study, we examine the ability of four food web metrics to predict the fragility to random species extinctions in 120 qualitative food webs. We show that three information-based indices out performed food web connectance in predicting fragility, with relative ascendency having the strongest relationship. Relative ascendency was a much stronger predictor of fragility than MacArthur’s stability metric, Average Mutual Information and connectance as it accounted for both the distribution and number of links between species. We also find that most qualitative food webs persist around a central tendency of relative ascendency.  相似文献   

16.

Background  

Very often genome-wide data analysis requires the interoperation of multiple databases and analytic tools. A large number of genome databases and bioinformatics applications are available through the web, but it is difficult to automate interoperation because: 1) the platforms on which the applications run are heterogeneous, 2) their web interface is not machine-friendly, 3) they use a non-standard format for data input and output, 4) they do not exploit standards to define application interface and message exchange, and 5) existing protocols for remote messaging are often not firewall-friendly. To overcome these issues, web services have emerged as a standard XML-based model for message exchange between heterogeneous applications. Web services engines have been developed to manage the configuration and execution of a web services workflow.  相似文献   

17.
The sterile male technique is a common method to assign paternity, widely adopted due to its relative simplicity and low cost. Male sterility is induced by exposure to sub lethal doses of chemosterilants or irradiation, the dosage of which has to be calibrated for every species to provide successful male sterilisation, without affecting male physiology and behaviour. While the physiological effects of sterilisation are usually assessed for each study, the behavioural ones are rarely analysed in detail. Using the orb web spider Argiope keyserlingi as a model we first tested (1) the validity of the thread assay, which simulates male courtship behaviour in a standardised context, as a proxy representing courtship on a female web. We then investigated (2) the effectiveness of male sterilisation via irradiation and (3) its consequences on male courtship behaviour. Our results validate the thread assay and the sterile male technique as legitimate tools for the study of male courtship behaviour and fertilisation success. We show that these techniques are time and cost effective and reduce undesirable variation, thereby creating opportunities to study and understand the mechanisms underlying sexual selection.  相似文献   

18.
Large quantities of data have been generated from multiple sources at exponential rates in the last few years. These data are generated at high velocity as real time and streaming data in variety of formats. These characteristics give rise to challenges in its modeling, computation, and processing. Hadoop MapReduce (MR) is a well known data-intensive distributed processing framework using the distributed file system (DFS) for Big Data. Current implementations of MR only support execution of a single algorithm in the entire Hadoop cluster. In this paper, we propose MapReducePack (MRPack), a variation of MR that supports execution of a set of related algorithms in a single MR job. We exploit the computational capability of a cluster by increasing the compute-intensiveness of MapReduce while maintaining its data-intensive approach. It uses the available computing resources by dynamically managing the task assignment and intermediate data. Intermediate data from multiple algorithms are managed using multi-key and skew mitigation strategies. The performance study of the proposed system shows that it is time, I/O, and memory efficient compared to the default MapReduce. The proposed approach reduces the execution time by 200% with an approximate 50% decrease in I/O cost. Complexity and qualitative results analysis shows significant performance improvement.  相似文献   

19.
Pencil beam algorithms are still considered as standard photon dose calculation methods in Radiotherapy treatment planning for many clinical applications. Despite their established role in radiotherapy planning their performance and clinical applicability has to be continuously adapted to evolving complex treatment techniques such as adaptive radiation therapy (ART). We herewith report on a new highly efficient version of a well-established pencil beam convolution algorithm which relies purely on measured input data. A method was developed that improves raytracing efficiency by exploiting the capability of modern CPU architecture for a runtime reduction. Since most of the current desktop computers provide more than one calculation unit we used symmetric multiprocessing extensively to parallelize the workload and thus decreasing the algorithmic runtime. To maximize the advantage of code parallelization, we present two implementation strategies – one for the dose calculation in inverse planning software, and one for traditional forward planning. As a result, we could achieve on a 16-core personal computer with AMD processors a superlinear speedup factor of approx. 18 for calculating the dose distribution of typical forward IMRT treatment plans.  相似文献   

20.
PurposeTo assess the effectiveness of SGRT in clinical applications through statistical process control (SPC).MethodsTaking the patients’ positioning through optical surface imaging (OSI) as a process, the average level of process execution was defined as the process mean. Setup errors detected by cone-beam computed tomography (CBCT) and OSI were extracted for head-and-neck cancer (HNC) and breast cancer patients. These data were used to construct individual and exponentially weighted moving average (EWMA) control charts to analyze outlier fractions and small process shifts from the process mean. Using the control charts and process capability indices derived from this process, the patient positioning-related OSI performance and setup error were analyzed for each patient.ResultsOutlier fractions and small shifts from the process mean that are indicative of setup errors were found to be widely prevalent, with the outliers randomly distributed between fractions. A systematic error of up to 1.6 mm between the OSI and CBCT results was observed in all directions, indicating a significantly degraded OSI performance. Adjusting this systematic error for each patient using setup errors of the first five fractions could effectively mitigate these effects. Process capability analysis following adjustment for systematic error indicated that OSI performance was acceptable (process capability index Cpk = 1.0) for HNC patients but unacceptable (Cpk < 0.75) for breast cancer patients.ConclusionSPC is a powerful tool for detecting the outlier fractions and process changes. Our application of SPC to patient-specific evaluations validated the suitability of OSI in clinical applications involving patient positioning.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号