This paper introduces a method realizing dynamic provisioning of services in a distributed environment. Depending on a particular state of infrastructure the call of a service can lead to a new instance in the infrastructure or to using an existing instance. Hence, the dynamic deployment allows optimized distribution of service instances within a certain infrastructure. The paper introduces a context model for services that are registered in a distributed runtime environment. Furthermore, algorithms are introduced determining the need for instantiation as well as the best location for deployment. Hence, the best location is determined by correlating the context model, the certain state of infrastructure as well as data transfer costs.
{"title":"Dynamic Service Provisioning for the Cloud","authors":"K. Görlach, F. Leymann","doi":"10.1109/SCC.2012.30","DOIUrl":"https://doi.org/10.1109/SCC.2012.30","url":null,"abstract":"This paper introduces a method realizing dynamic provisioning of services in a distributed environment. Depending on a particular state of infrastructure the call of a service can lead to a new instance in the infrastructure or to using an existing instance. Hence, the dynamic deployment allows optimized distribution of service instances within a certain infrastructure. The paper introduces a context model for services that are registered in a distributed runtime environment. Furthermore, algorithms are introduced determining the need for instantiation as well as the best location for deployment. Hence, the best location is determined by correlating the context model, the certain state of infrastructure as well as data transfer costs.","PeriodicalId":178841,"journal":{"name":"2012 IEEE Ninth International Conference on Services Computing","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125392440","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Bipin B. Nandi, A. Banerjee, Sasthi C. Ghosh, N. Banerjee
Virtual machine (VM) placement for Datacenter (DC) consolidation is a challenging problem, particularly in the face of VM workload fluctuation. In this paper, we present a stochastic model for optimization of DC consolidation and propose intelligent strategies for statistical VM multiplexing on physical machines (PMs) to ensure optimal use of hardware resources, while providing a service guarantee. We have provided an optimal strategy by modeling and solving the problem as a stochastic integer programming problem followed by a more scalable strategy based on a greedy heuristic. Extensive simulation based experimental results show that the strategies are more efficient in resource utilization while providing bounded service guarantees, than the traditional way of VM placement without any consideration to workload fluctuation.
{"title":"Stochastic VM Multiplexing for Datacenter Consolidation","authors":"Bipin B. Nandi, A. Banerjee, Sasthi C. Ghosh, N. Banerjee","doi":"10.1109/SCC.2012.94","DOIUrl":"https://doi.org/10.1109/SCC.2012.94","url":null,"abstract":"Virtual machine (VM) placement for Datacenter (DC) consolidation is a challenging problem, particularly in the face of VM workload fluctuation. In this paper, we present a stochastic model for optimization of DC consolidation and propose intelligent strategies for statistical VM multiplexing on physical machines (PMs) to ensure optimal use of hardware resources, while providing a service guarantee. We have provided an optimal strategy by modeling and solving the problem as a stochastic integer programming problem followed by a more scalable strategy based on a greedy heuristic. Extensive simulation based experimental results show that the strategies are more efficient in resource utilization while providing bounded service guarantees, than the traditional way of VM placement without any consideration to workload fluctuation.","PeriodicalId":178841,"journal":{"name":"2012 IEEE Ninth International Conference on Services Computing","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126739609","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Recently many organizations have accumulated data from such various sources as web and network sensors and constructed large-scale archives. Some would like to publish their archives to public to facilitate the activities of other organizations, but the scale of the archives causes problems. Therefore, we propose the concept of data-intensive services, which publish large-scale archives. We show the architecture for data-intensive services and focus on the following fundamental functional properties: 1) enhancing search, 2) preprocessing, 3) and asynchronous transfer. We also developed a reference implementation of a framework for data-intensive services and applied it to a web archive that contains about 2 billion documents and greatly improved the access performance to the web archive at small development cost.
{"title":"Data-Intensive Services for Large-Scale Archive Access","authors":"Masahiro Tanaka, Yohei Murakami, K. Zettsu","doi":"10.1109/SCC.2012.75","DOIUrl":"https://doi.org/10.1109/SCC.2012.75","url":null,"abstract":"Recently many organizations have accumulated data from such various sources as web and network sensors and constructed large-scale archives. Some would like to publish their archives to public to facilitate the activities of other organizations, but the scale of the archives causes problems. Therefore, we propose the concept of data-intensive services, which publish large-scale archives. We show the architecture for data-intensive services and focus on the following fundamental functional properties: 1) enhancing search, 2) preprocessing, 3) and asynchronous transfer. We also developed a reference implementation of a framework for data-intensive services and applied it to a web archive that contains about 2 billion documents and greatly improved the access performance to the web archive at small development cost.","PeriodicalId":178841,"journal":{"name":"2012 IEEE Ninth International Conference on Services Computing","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116243147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As Cloud Computing has become more and more popular, various Cloud Computing architectures or infrastructure have been defined, given their specific circumstances for the applications. However, to effectively achieve the potential of cloud computing, there is need for the definition of system architecture of the software systems involved in the delivery of cloud computing, so that it can be used as a reference for the architects or software engineering. In this paper, reference architecture of Cloud Computing is proposed. Its objective and principles are illustrated. And case studies of a SaaS, PaaS platform architecture instantiated from CCRA are given.
{"title":"CCRA: Cloud Computing Reference Architecture","authors":"Jing Liu, Liang-Jie Zhang, Bo Hu, K. He","doi":"10.1109/SCC.2012.110","DOIUrl":"https://doi.org/10.1109/SCC.2012.110","url":null,"abstract":"As Cloud Computing has become more and more popular, various Cloud Computing architectures or infrastructure have been defined, given their specific circumstances for the applications. However, to effectively achieve the potential of cloud computing, there is need for the definition of system architecture of the software systems involved in the delivery of cloud computing, so that it can be used as a reference for the architects or software engineering. In this paper, reference architecture of Cloud Computing is proposed. Its objective and principles are illustrated. And case studies of a SaaS, PaaS platform architecture instantiated from CCRA are given.","PeriodicalId":178841,"journal":{"name":"2012 IEEE Ninth International Conference on Services Computing","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122368176","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Stock market is an important and active part of nowadays financial markets. Stock time series volatility analysis is regarded as one of the most challenging time series forecasting due to the hard-to-predict volatility observed in worldwide stock markets. In this paper we argue that the stock market state is dynamic and invisible but it will be influenced by some visible stock market information. Existing research on financial time series analysis and stock market volatility prediction can be classified into two categories: in depth study of one market factor on the stock market volatility prediction or prediction by combining historical price fluctuations with either trading volume or news. In this paper we present a service-oriented multi-kernel based learning framework (MKL) for stock volatility analysis. Our MKL service framework promotes a two-tier learning architecture. In the top tier, we develop a suite of data preparation and data transformation techniques to provide a source-specific modeling, which transforms and normalizes a source specific input dataset into the MKL ready data representation. Then we apply data alignment techniques to prepare the datasets from multiple information sources based on the classification model we choose for cross-source correlation analysis. In the next tier, we develop model integration methods to perform three analytic tasks: (i) building one sub-kernel per source, (ii) learning and tuning the weights for sub-kernels through weight adjustment methods and (iii) performing multi-kernel based cross-correlation analysis of market volatility. To validate the effectiveness of our service oriented MKL approach, we performed experiments on HKEx 2001 stock market datasets with three important market information sources: historical prices, trading volumes and stock related news articles. Our experiments show that 1) multi-kernel learning method has a higher degree of accuracy and a lower degree of false prediction, compared to existing single kernel methods; and 2) integrating both news and trading volume data with historical stock price information can significantly improve the effectiveness of stock market volatility prediction, compared to many existing prediction methods.
{"title":"Stock Market Volatility Prediction: A Service-Oriented Multi-kernel Learning Approach","authors":"Feng Wang, Ling Liu, Chenxiao Dou","doi":"10.1109/SCC.2012.35","DOIUrl":"https://doi.org/10.1109/SCC.2012.35","url":null,"abstract":"Stock market is an important and active part of nowadays financial markets. Stock time series volatility analysis is regarded as one of the most challenging time series forecasting due to the hard-to-predict volatility observed in worldwide stock markets. In this paper we argue that the stock market state is dynamic and invisible but it will be influenced by some visible stock market information. Existing research on financial time series analysis and stock market volatility prediction can be classified into two categories: in depth study of one market factor on the stock market volatility prediction or prediction by combining historical price fluctuations with either trading volume or news. In this paper we present a service-oriented multi-kernel based learning framework (MKL) for stock volatility analysis. Our MKL service framework promotes a two-tier learning architecture. In the top tier, we develop a suite of data preparation and data transformation techniques to provide a source-specific modeling, which transforms and normalizes a source specific input dataset into the MKL ready data representation. Then we apply data alignment techniques to prepare the datasets from multiple information sources based on the classification model we choose for cross-source correlation analysis. In the next tier, we develop model integration methods to perform three analytic tasks: (i) building one sub-kernel per source, (ii) learning and tuning the weights for sub-kernels through weight adjustment methods and (iii) performing multi-kernel based cross-correlation analysis of market volatility. To validate the effectiveness of our service oriented MKL approach, we performed experiments on HKEx 2001 stock market datasets with three important market information sources: historical prices, trading volumes and stock related news articles. Our experiments show that 1) multi-kernel learning method has a higher degree of accuracy and a lower degree of false prediction, compared to existing single kernel methods; and 2) integrating both news and trading volume data with historical stock price information can significantly improve the effectiveness of stock market volatility prediction, compared to many existing prediction methods.","PeriodicalId":178841,"journal":{"name":"2012 IEEE Ninth International Conference on Services Computing","volume":"10 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122208751","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yexi Jiang, Chang-Shing Perng, Tao Li, Rong N. Chang
The popularity of cloud service spurs the increasing demands of cloud resources to the cloud service providers. Along with the new business opportunities, the pay-as-you-go model drastically changes the usage pattern and brings technology challenges to effective capacity planning. In this paper, we propose a new method for cloud capacity planning with the goal of fully utilizing the physical resources, as we believe this is one of the emerging problems for cloud providers. To solve this problem, we present an integrated system with intelligent cloud capacity prediction. Considering the unique characteristics of the cloud service that virtual machines are provisioned and de-provisioned frequently to meet the business needs, we propose an asymmetric and heterogeneous measure for modeling the over-estimation, and under-estimation of the capacity. To accurately forecast the capacity, we first divide the change of cloud capacity demand into provisioning and de-provisioning components, and then estimate the individual components respectively. The future provisioning demand is predicted by an ensemble time-series prediction method, while the future de-provisioning is inferred based on the life span distribution and the number of active virtual machines. Our proposed solution is simple and computational efficient, which make it practical for development and deployment. Our solution also has the advantages for generating interpretable predictions. The experimental results on the IBM Smart Cloud Enterprise trace data demonstrate the effectiveness, accuracy and efficiency of our solution.
{"title":"Self-Adaptive Cloud Capacity Planning","authors":"Yexi Jiang, Chang-Shing Perng, Tao Li, Rong N. Chang","doi":"10.1109/SCC.2012.8","DOIUrl":"https://doi.org/10.1109/SCC.2012.8","url":null,"abstract":"The popularity of cloud service spurs the increasing demands of cloud resources to the cloud service providers. Along with the new business opportunities, the pay-as-you-go model drastically changes the usage pattern and brings technology challenges to effective capacity planning. In this paper, we propose a new method for cloud capacity planning with the goal of fully utilizing the physical resources, as we believe this is one of the emerging problems for cloud providers. To solve this problem, we present an integrated system with intelligent cloud capacity prediction. Considering the unique characteristics of the cloud service that virtual machines are provisioned and de-provisioned frequently to meet the business needs, we propose an asymmetric and heterogeneous measure for modeling the over-estimation, and under-estimation of the capacity. To accurately forecast the capacity, we first divide the change of cloud capacity demand into provisioning and de-provisioning components, and then estimate the individual components respectively. The future provisioning demand is predicted by an ensemble time-series prediction method, while the future de-provisioning is inferred based on the life span distribution and the number of active virtual machines. Our proposed solution is simple and computational efficient, which make it practical for development and deployment. Our solution also has the advantages for generating interpretable predictions. The experimental results on the IBM Smart Cloud Enterprise trace data demonstrate the effectiveness, accuracy and efficiency of our solution.","PeriodicalId":178841,"journal":{"name":"2012 IEEE Ninth International Conference on Services Computing","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128371282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper addresses the issue of selecting Web services residing in a community. Since these Web services have similar functionalities, this selection depends on their Quality of Service (QoS). Existing approaches only consider the satisfaction of users' requirements and neglect the satisfaction of Web services' requirements and the community to which they belong. This paper proposes an approach of selecting Web services based on the satisfaction of all three parties - user, Web service, and community. The approach consists of first, formalizing the selection process and then, using integer programming to define a score function, which can be maximized to find the best selection based on three satisfaction factors. Experiments using real Web services and measurements are conducted to demonstrate the influences of the approach.
{"title":"On the Analysis of Satisfaction for Web Services Selection","authors":"E. Lim, Philippe Thiran, Z. Maamar, J. Bentahar","doi":"10.1109/SCC.2012.66","DOIUrl":"https://doi.org/10.1109/SCC.2012.66","url":null,"abstract":"This paper addresses the issue of selecting Web services residing in a community. Since these Web services have similar functionalities, this selection depends on their Quality of Service (QoS). Existing approaches only consider the satisfaction of users' requirements and neglect the satisfaction of Web services' requirements and the community to which they belong. This paper proposes an approach of selecting Web services based on the satisfaction of all three parties - user, Web service, and community. The approach consists of first, formalizing the selection process and then, using integer programming to define a score function, which can be maximized to find the best selection based on three satisfaction factors. Experiments using real Web services and measurements are conducted to demonstrate the influences of the approach.","PeriodicalId":178841,"journal":{"name":"2012 IEEE Ninth International Conference on Services Computing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130206898","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Y. Maurel, Stéphanie Chollet, V. Lestideau, J. Bardin, P. Lalanda, A. Bottaro
The ability to react quickly to unpredictable changes in the environment is a key requirement in pervasive computing. This paper presents fANFARE, a framework for the autonomic management of service-oriented applications in pervasive environments. Specifically, it focuses on the configuration and optimization of pervasive applications deployed on OSGi platforms. We propose to handle runtime administration through a hierarchy of autonomic managers, that is a platform manager and a number of application managers and dependency managers. Our approach has been implemented and validated on pervasive use cases within the MEDICAL project funded by the French Ministry of Industry.
{"title":"fANFARE: Autonomic Framework for Service-Based Pervasive Environment","authors":"Y. Maurel, Stéphanie Chollet, V. Lestideau, J. Bardin, P. Lalanda, A. Bottaro","doi":"10.1109/SCC.2012.7","DOIUrl":"https://doi.org/10.1109/SCC.2012.7","url":null,"abstract":"The ability to react quickly to unpredictable changes in the environment is a key requirement in pervasive computing. This paper presents fANFARE, a framework for the autonomic management of service-oriented applications in pervasive environments. Specifically, it focuses on the configuration and optimization of pervasive applications deployed on OSGi platforms. We propose to handle runtime administration through a hierarchy of autonomic managers, that is a platform manager and a number of application managers and dependency managers. Our approach has been implemented and validated on pervasive use cases within the MEDICAL project funded by the French Ministry of Industry.","PeriodicalId":178841,"journal":{"name":"2012 IEEE Ninth International Conference on Services Computing","volume":"321 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126587838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
For Component Based Software Engineering (CBSE), an application is a strongly structured and rigid assembly of components. Conversely, Service Oriented Computing (SOC) is very flexible and is a good candidate for supporting dynamic applications. Unfortunately dynamic applications are software applications and as such they need to be clearly structured and managed (as with CBSE), and they need flexibility and dynamism as with SOC. No platform today satisfies both needs. This paper presents the Component-Service model that combines well controlled structure and dynamism, and its implementation into the Apam component-service platform.
{"title":"Reconciling Components and Services: The Apam Component-Service Platform","authors":"J. Estublier, Germán Vega","doi":"10.1109/SCC.2012.14","DOIUrl":"https://doi.org/10.1109/SCC.2012.14","url":null,"abstract":"For Component Based Software Engineering (CBSE), an application is a strongly structured and rigid assembly of components. Conversely, Service Oriented Computing (SOC) is very flexible and is a good candidate for supporting dynamic applications. Unfortunately dynamic applications are software applications and as such they need to be clearly structured and managed (as with CBSE), and they need flexibility and dynamism as with SOC. No platform today satisfies both needs. This paper presents the Component-Service model that combines well controlled structure and dynamism, and its implementation into the Apam component-service platform.","PeriodicalId":178841,"journal":{"name":"2012 IEEE Ninth International Conference on Services Computing","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127404603","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Daxiang Zhao, Shijun Liu, Lei Wu, Rui Wang, Xiangxu Meng
Software-as-a-service emerges as a new delivery model for the software development and applications. In multi-tenancy applications, tenants can customize and adjust the corresponding services shared by tenants based on their business requirements with strong elasticity and expansibility. As services do not exist independently, the resolving of service dependencies in practical multi-tenancy applications is of great importance. In this paper, we propose an Extended Dependency-aware Hierarchical Service Model (EDHSM) to visually describe hierarchical services and service dependencies. We employ a directed hypergraph to represent the model and resolve service dependencies. Moreover, we apply the service dependencies resolving to dynamic service deployment in optimal placement of tenants and online service migration based on load monitoring for multi-tenancy applications. So application systems can be built quickly and operated stably at a low maintaining cost.
{"title":"Hypergraph-Based Service Dependency Resolving and Its Applications","authors":"Daxiang Zhao, Shijun Liu, Lei Wu, Rui Wang, Xiangxu Meng","doi":"10.1109/SCC.2012.25","DOIUrl":"https://doi.org/10.1109/SCC.2012.25","url":null,"abstract":"Software-as-a-service emerges as a new delivery model for the software development and applications. In multi-tenancy applications, tenants can customize and adjust the corresponding services shared by tenants based on their business requirements with strong elasticity and expansibility. As services do not exist independently, the resolving of service dependencies in practical multi-tenancy applications is of great importance. In this paper, we propose an Extended Dependency-aware Hierarchical Service Model (EDHSM) to visually describe hierarchical services and service dependencies. We employ a directed hypergraph to represent the model and resolve service dependencies. Moreover, we apply the service dependencies resolving to dynamic service deployment in optimal placement of tenants and online service migration based on load monitoring for multi-tenancy applications. So application systems can be built quickly and operated stably at a low maintaining cost.","PeriodicalId":178841,"journal":{"name":"2012 IEEE Ninth International Conference on Services Computing","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2012-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127365077","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：