Pub Date : 1998-12-14DOI: 10.1109/ICPADS.1998.741129
Y. Kamizuru, Y. Amagai
Distributed Internet Route Eye (DIRE) is a distributed agent architecture for inter-autonomous system (AS) route monitoring. In this architecture, monitor agents (MAs) record detailed BGP route updates from border gateway protocol (BGP) routers, and operator agents (OAs) query multiple MAs for specific information and infer the source and cause of the problems. Currently, MA implementation is near completion and actual route transitions can be monitored. The potential for flap source inference by OAs is examined using actual route transition data monitored by MAs. This system is capable of providing faster, easier access to more detailed information for coping with routing instabilities.
分布式Internet路由眼(Distributed Internet Route Eye，简称DIRE)是一种用于自治系统间路由监控的分布式代理体系结构。在该体系结构中，监控代理(monitor agent)详细记录来自BGP路由器的BGP路由更新，而运营商代理(operator agent)则通过查询多个MAs来获取特定信息，并推断出问题的来源和原因。目前，MA的实现已经接近完成，并且可以监控实际的路由转换。利用MAs监测的实际航路转换数据，检验了oa对襟翼源推断的可能性。该系统能够提供更快，更容易地访问更详细的信息，以应对路由不稳定。
{"title":"Distributed Inter-AS route monitor-Distributed Internet Route Eye (DIRE)","authors":"Y. Kamizuru, Y. Amagai","doi":"10.1109/ICPADS.1998.741129","DOIUrl":"https://doi.org/10.1109/ICPADS.1998.741129","url":null,"abstract":"Distributed Internet Route Eye (DIRE) is a distributed agent architecture for inter-autonomous system (AS) route monitoring. In this architecture, monitor agents (MAs) record detailed BGP route updates from border gateway protocol (BGP) routers, and operator agents (OAs) query multiple MAs for specific information and infer the source and cause of the problems. Currently, MA implementation is near completion and actual route transitions can be monitored. The potential for flap source inference by OAs is examined using actual route transition data monitored by MAs. This system is capable of providing faster, easier access to more detailed information for coping with routing instabilities.","PeriodicalId":226947,"journal":{"name":"Proceedings 1998 International Conference on Parallel and Distributed Systems (Cat. No.98TB100250)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1998-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115555396","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}
Pub Date : 1998-12-14DOI: 10.1109/ICPADS.1998.741037
Keith K. S. Lee, Y. Chin
The goals of this paper are to revise previous replication strategies and propose a new hierarchy structure, client-agent-server architecture, to fit a mobile computing system that has more resource constraints than a conventional distributed database system has. This three-tier agent-based mobile computing system frees the limitations on time and whereabouts of computing. Moreover, a new replication strategy, three-copy replication that is based on an optimistic replication strategy, is proposed to deal with the problems of mobility management and disconnection management in a mobile computing system. It allows mobile clients to read and update the database while they are disconnected from the network, if the mobile copy is available. In addition, the implementation issues of this new replication strategy are also provided.
{"title":"A new replication strategy for unforeseeable disconnection under agent-based mobile computing system","authors":"Keith K. S. Lee, Y. Chin","doi":"10.1109/ICPADS.1998.741037","DOIUrl":"https://doi.org/10.1109/ICPADS.1998.741037","url":null,"abstract":"The goals of this paper are to revise previous replication strategies and propose a new hierarchy structure, client-agent-server architecture, to fit a mobile computing system that has more resource constraints than a conventional distributed database system has. This three-tier agent-based mobile computing system frees the limitations on time and whereabouts of computing. Moreover, a new replication strategy, three-copy replication that is based on an optimistic replication strategy, is proposed to deal with the problems of mobility management and disconnection management in a mobile computing system. It allows mobile clients to read and update the database while they are disconnected from the network, if the mobile copy is available. In addition, the implementation issues of this new replication strategy are also provided.","PeriodicalId":226947,"journal":{"name":"Proceedings 1998 International Conference on Parallel and Distributed Systems (Cat. No.98TB100250)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1998-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114528964","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}
Pub Date : 1998-12-14DOI: 10.1109/ICPADS.1998.741118
T. Hardjono, B. Cain
The paper contributes a new architecture for secure and scalable inter-domain group key management for N-to-N (conference) IP multicast. The architecture views the multicast routing infrastructure from the key management plane, and logically divides it into two general types of regions for key management to achieve scalability. The work extends the centralized solution of (Wong et al., 1998) into a distributed key management scheme suitable for inter-domain multicast group key management. Methods for initiating new multicast groups, as well as for members joining and leaving, are presented. The paper also considers two general types of IP-multicast that need to be made secure if multicast is to be one of the vehicles for future wide-scale delivery of voice, video and text over the Internet.
本文提出了一种安全、可扩展的N-to-N(会议)IP组播域间组密钥管理新体系结构。该体系结构从密钥管理平面来看待组播路由基础设施，并在逻辑上将其划分为两种一般类型的区域进行密钥管理，以实现可扩展性。该工作将(Wong et al.， 1998)的集中式解决方案扩展为适用于域间多播组密钥管理的分布式密钥管理方案。提出了发起新组播组以及成员加入和离开组播组的方法。本文还考虑了两种一般类型的ip组播，如果组播要成为未来在互联网上大规模传输语音、视频和文本的载体之一，则需要使其安全。
{"title":"Secure and scalable inter-domain group key management for N-to-N multicast","authors":"T. Hardjono, B. Cain","doi":"10.1109/ICPADS.1998.741118","DOIUrl":"https://doi.org/10.1109/ICPADS.1998.741118","url":null,"abstract":"The paper contributes a new architecture for secure and scalable inter-domain group key management for N-to-N (conference) IP multicast. The architecture views the multicast routing infrastructure from the key management plane, and logically divides it into two general types of regions for key management to achieve scalability. The work extends the centralized solution of (Wong et al., 1998) into a distributed key management scheme suitable for inter-domain multicast group key management. Methods for initiating new multicast groups, as well as for members joining and leaving, are presented. The paper also considers two general types of IP-multicast that need to be made secure if multicast is to be one of the vehicles for future wide-scale delivery of voice, video and text over the Internet.","PeriodicalId":226947,"journal":{"name":"Proceedings 1998 International Conference on Parallel and Distributed Systems (Cat. No.98TB100250)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1998-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115462186","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}
Pub Date : 1998-12-14DOI: 10.1109/ICPADS.1998.741026
T. Hong, Jyh-Jong Lee
We first propose a modified backpropagation learning algorithm that incrementally decreases the error threshold by half in order to process training instances with large weight changes as quickly as possible. This modified backpropagation learning algorithm is then parallelized using the single-channel broadcast communication model to n processors, where n is the number of training instances. Finally, the parallel backpropagation learning algorithm is modified for execution on a bounded number of processors to cope with real-world conditions.
{"title":"Parallel neural learning by iteratively adjusting error thresholds","authors":"T. Hong, Jyh-Jong Lee","doi":"10.1109/ICPADS.1998.741026","DOIUrl":"https://doi.org/10.1109/ICPADS.1998.741026","url":null,"abstract":"We first propose a modified backpropagation learning algorithm that incrementally decreases the error threshold by half in order to process training instances with large weight changes as quickly as possible. This modified backpropagation learning algorithm is then parallelized using the single-channel broadcast communication model to n processors, where n is the number of training instances. Finally, the parallel backpropagation learning algorithm is modified for execution on a bounded number of processors to cope with real-world conditions.","PeriodicalId":226947,"journal":{"name":"Proceedings 1998 International Conference on Parallel and Distributed Systems (Cat. No.98TB100250)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1998-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123075381","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}
Pub Date : 1998-12-14DOI: 10.1109/ICPADS.1998.741142
A. Bader, T. Elrad
In this paper we present a concurrent object-oriented model in which a concurrent object, which represents a shared resource abstraction in our model, is decomposed into a hierarchy of architectural abstractions: a shared data abstraction, a synchronization abstraction, and a scheduling abstraction. It will be shown that the separation of concerns among the three major components of the concurrent objects avoids many of the conceptual difficulties that arise when integrating concurrency into the object-oriented paradigm. The proposed model presents a formal methodology for the architectural design and specification of the concurrent object-oriented software systems. The notion of scheduling policy inheritance in our model facilitates the process of engineering adaptability in the development of the intelligent reactive/adaptive systems.
{"title":"The Adaptive Arena: language constructs and architectural abstractions for concurrent object-oriented systems","authors":"A. Bader, T. Elrad","doi":"10.1109/ICPADS.1998.741142","DOIUrl":"https://doi.org/10.1109/ICPADS.1998.741142","url":null,"abstract":"In this paper we present a concurrent object-oriented model in which a concurrent object, which represents a shared resource abstraction in our model, is decomposed into a hierarchy of architectural abstractions: a shared data abstraction, a synchronization abstraction, and a scheduling abstraction. It will be shown that the separation of concerns among the three major components of the concurrent objects avoids many of the conceptual difficulties that arise when integrating concurrency into the object-oriented paradigm. The proposed model presents a formal methodology for the architectural design and specification of the concurrent object-oriented software systems. The notion of scheduling policy inheritance in our model facilitates the process of engineering adaptability in the development of the intelligent reactive/adaptive systems.","PeriodicalId":226947,"journal":{"name":"Proceedings 1998 International Conference on Parallel and Distributed Systems (Cat. No.98TB100250)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1998-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129221038","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}
Pub Date : 1998-12-14DOI: 10.1109/ICPADS.1998.741027
W. Shih, Chung-Der Lin, Yar-Wen Chang, Jenq-Kuen Lee
Gang scheduling has been shown to be an effective job scheduling policy for parallel computers that combines elements of space sharing and time sharing. We propose new policies to enable gang scheduling to adapt to environments with real-time constraints. Our work, to our best knowledge, is the first work to attempt to address the real-time aspects of gang scheduling. Our system guided by a metric, called "task utilization workload", can schedule both real-time and non-real-time tasks at the same time. We report simulation results with a family of scheduling algorithms based on our proposed metric. Our scheme is designed to be a practical scheme to be used for large scale industrial and commercial parallel systems. Preliminary simulation results also show that our proposed policy is an effective scheme to perform real-time scheduling, while scheduling non-real-time jobs with fairness and good throughput.
{"title":"Real-time gang schedulings with workload models for parallel computers","authors":"W. Shih, Chung-Der Lin, Yar-Wen Chang, Jenq-Kuen Lee","doi":"10.1109/ICPADS.1998.741027","DOIUrl":"https://doi.org/10.1109/ICPADS.1998.741027","url":null,"abstract":"Gang scheduling has been shown to be an effective job scheduling policy for parallel computers that combines elements of space sharing and time sharing. We propose new policies to enable gang scheduling to adapt to environments with real-time constraints. Our work, to our best knowledge, is the first work to attempt to address the real-time aspects of gang scheduling. Our system guided by a metric, called \"task utilization workload\", can schedule both real-time and non-real-time tasks at the same time. We report simulation results with a family of scheduling algorithms based on our proposed metric. Our scheme is designed to be a practical scheme to be used for large scale industrial and commercial parallel systems. Preliminary simulation results also show that our proposed policy is an effective scheme to perform real-time scheduling, while scheduling non-real-time jobs with fairness and good throughput.","PeriodicalId":226947,"journal":{"name":"Proceedings 1998 International Conference on Parallel and Distributed Systems (Cat. No.98TB100250)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1998-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129422167","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}
Pub Date : 1998-12-14DOI: 10.1109/ICPADS.1998.741035
Y. Tseng, Cheng-Chung Tan
Incorporating mobile components into a distributed system has posed new challenges to the design of distributed computation. This paper studies a fundamental problem in distributed computing, the termination detection problem, in a mobile environment. Two types of termination detection protocols already exist: the diffusion-based schemes and the weight-throwing schemes, that are designed for traditional static distributed systems. We propose a hybrid scheme by combining these two protocols together. The scheme can better exploit the communication hierarchy (in terms of wired and wireless bandwidths) and can pave the gaps of computation and communication capability between static and mobile hosts, thus more scalable to larger distributed systems. Simulation results are presented, which show the advantage of the hybrid scheme over existing schemes.
{"title":"On termination detection protocols in a mobile distributed computing environment","authors":"Y. Tseng, Cheng-Chung Tan","doi":"10.1109/ICPADS.1998.741035","DOIUrl":"https://doi.org/10.1109/ICPADS.1998.741035","url":null,"abstract":"Incorporating mobile components into a distributed system has posed new challenges to the design of distributed computation. This paper studies a fundamental problem in distributed computing, the termination detection problem, in a mobile environment. Two types of termination detection protocols already exist: the diffusion-based schemes and the weight-throwing schemes, that are designed for traditional static distributed systems. We propose a hybrid scheme by combining these two protocols together. The scheme can better exploit the communication hierarchy (in terms of wired and wireless bandwidths) and can pave the gaps of computation and communication capability between static and mobile hosts, thus more scalable to larger distributed systems. Simulation results are presented, which show the advantage of the hybrid scheme over existing schemes.","PeriodicalId":226947,"journal":{"name":"Proceedings 1998 International Conference on Parallel and Distributed Systems (Cat. No.98TB100250)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1998-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127037273","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}
Pub Date : 1998-12-14DOI: 10.1109/ICPADS.1998.741130
Yui-Wen Horng, Wen Lin, Hsing Mei
As the application of the World Wide Web grows, technologies for reducing Web latency are becoming important. Proxy caching is one of the most effective and widely implemented mechanisms. By prefetching documents on the proxy server, a better cache hit rate can be achieved and the clients experience a faster response time. In this thesis, we propose a hybrid prefetching technique, combined with different constraints, which could increase cache hit rate. Under this new prefetching scheme, the additional traffic is controllable. We also show that, when prefetching techniques are used in the proxy server, separated cache management is better if the whole cache size is small. All of the mechanisms are experimented by trace driven simulation.
{"title":"Hybrid prefetching for WWW proxy servers","authors":"Yui-Wen Horng, Wen Lin, Hsing Mei","doi":"10.1109/ICPADS.1998.741130","DOIUrl":"https://doi.org/10.1109/ICPADS.1998.741130","url":null,"abstract":"As the application of the World Wide Web grows, technologies for reducing Web latency are becoming important. Proxy caching is one of the most effective and widely implemented mechanisms. By prefetching documents on the proxy server, a better cache hit rate can be achieved and the clients experience a faster response time. In this thesis, we propose a hybrid prefetching technique, combined with different constraints, which could increase cache hit rate. Under this new prefetching scheme, the additional traffic is controllable. We also show that, when prefetching techniques are used in the proxy server, separated cache management is better if the whole cache size is small. All of the mechanisms are experimented by trace driven simulation.","PeriodicalId":226947,"journal":{"name":"Proceedings 1998 International Conference on Parallel and Distributed Systems (Cat. No.98TB100250)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1998-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126798360","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}
Pub Date : 1998-12-14DOI: 10.1109/ICPADS.1998.741110
A. Cortés, A. Ripoll, M. A. Senar, F. Cedó, E. Luque
We present a new fully distributed dynamic load balancing algorithm called DASUD (Diffusion Algorithm Searching Unbalanced Domains). Since DASUD is iterative and runs in an asynchronous way, a mathematical model that describes DASUD behaviour has been proposed and has been used to prove DASUD's convergence. DASUD has been evaluated by comparison with another well known strategy from the literature, namely, the SID (Sender Initiated Diffusion) algorithm. The comparison was carried out by considering a large set of load distributions which were applied to ring, torus and hypercube topologies, and the number of processors ranged from 8 to 128. From these experiments we have observed that DASUD outperforms the SID strategy as it provides the best trade-off between the global balance degree obtained at the final state and the number of iterations required to reach such a state.
{"title":"On the stability of a distributed dynamic load balancing algorithm","authors":"A. Cortés, A. Ripoll, M. A. Senar, F. Cedó, E. Luque","doi":"10.1109/ICPADS.1998.741110","DOIUrl":"https://doi.org/10.1109/ICPADS.1998.741110","url":null,"abstract":"We present a new fully distributed dynamic load balancing algorithm called DASUD (Diffusion Algorithm Searching Unbalanced Domains). Since DASUD is iterative and runs in an asynchronous way, a mathematical model that describes DASUD behaviour has been proposed and has been used to prove DASUD's convergence. DASUD has been evaluated by comparison with another well known strategy from the literature, namely, the SID (Sender Initiated Diffusion) algorithm. The comparison was carried out by considering a large set of load distributions which were applied to ring, torus and hypercube topologies, and the number of processors ranged from 8 to 128. From these experiments we have observed that DASUD outperforms the SID strategy as it provides the best trade-off between the global balance degree obtained at the final state and the number of iterations required to reach such a state.","PeriodicalId":226947,"journal":{"name":"Proceedings 1998 International Conference on Parallel and Distributed Systems (Cat. No.98TB100250)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1998-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128763259","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}
Pub Date : 1998-12-14DOI: 10.1109/ICPADS.1998.741053
Guan-Chi Chen, Suh-Yin Lee
There are more and more information services provided on the wireless network. Due to long network delay of wireless link, the transaction will be a long lived transaction. In such a situation, the occurrence of handoff is inevitable, and thus a wireless link held by the mobile unit crossing the cell boundaries might be forced to terminate. It is undesirable that an active transaction is forced to terminate. We propose two guard channel schemes (GCS), static and dynamic, to reduce the probability of forced termination of transactions. In dynamic GCS, the number of reserved channels of a base station is dynamically assigned according to the number of transaction calls which may handoff to this cell while the number of guard channels is fixed in the static GCS. An analytic model based on Markov chain is derived to evaluate the system performance. The correctness of this model is verified by a simulation program. The experimental results show that a significant improvement is achieved by using the dynamic GCS.
{"title":"Modeling the static and dynamic guard channel schemes for mobile transactions","authors":"Guan-Chi Chen, Suh-Yin Lee","doi":"10.1109/ICPADS.1998.741053","DOIUrl":"https://doi.org/10.1109/ICPADS.1998.741053","url":null,"abstract":"There are more and more information services provided on the wireless network. Due to long network delay of wireless link, the transaction will be a long lived transaction. In such a situation, the occurrence of handoff is inevitable, and thus a wireless link held by the mobile unit crossing the cell boundaries might be forced to terminate. It is undesirable that an active transaction is forced to terminate. We propose two guard channel schemes (GCS), static and dynamic, to reduce the probability of forced termination of transactions. In dynamic GCS, the number of reserved channels of a base station is dynamically assigned according to the number of transaction calls which may handoff to this cell while the number of guard channels is fixed in the static GCS. An analytic model based on Markov chain is derived to evaluate the system performance. The correctness of this model is verified by a simulation program. The experimental results show that a significant improvement is achieved by using the dynamic GCS.","PeriodicalId":226947,"journal":{"name":"Proceedings 1998 International Conference on Parallel and Distributed Systems (Cat. No.98TB100250)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1998-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122345293","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：