Pub Date : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137762
Y. Sit, Cho-Li Wang, F. Lau
Cluster-based web server is a popular solution to meet the demand of the ever-growing web traffic. However existing approaches suffer from several limitations to achieve this. Dispatcher-based systems either can achieve only coarse-grained load balancing or would introduce heavy load to the dispatcher Mechanisms like cooperative caching consume much network resources when transferring large cache objects. In this paper, we present a new network support mechanism, called Socket Cloning (SC), in which an opened socket can be migrated efficiently between cluster nodes. With SC, the processing of HTTP requests can be moved to the node that has a cached copy of the requested document, thus bypassing any object transfer between peer servers. A prototype has been implemented and tests show that SC incurs less overhead than all the mentioned approaches. In trace-driven benchmark tests, our system outperforms these approaches by more than 30% with a cluster of twelve web server nodes.
{"title":"Socket cloning for cluster-based web servers","authors":"Y. Sit, Cho-Li Wang, F. Lau","doi":"10.1109/CLUSTR.2002.1137762","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137762","url":null,"abstract":"Cluster-based web server is a popular solution to meet the demand of the ever-growing web traffic. However existing approaches suffer from several limitations to achieve this. Dispatcher-based systems either can achieve only coarse-grained load balancing or would introduce heavy load to the dispatcher Mechanisms like cooperative caching consume much network resources when transferring large cache objects. In this paper, we present a new network support mechanism, called Socket Cloning (SC), in which an opened socket can be migrated efficiently between cluster nodes. With SC, the processing of HTTP requests can be moved to the node that has a cached copy of the requested document, thus bypassing any object transfer between peer servers. A prototype has been implemented and tests show that SC incurs less overhead than all the mentioned approaches. In trace-driven benchmark tests, our system outperforms these approaches by more than 30% with a cluster of twelve web server nodes.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":"14 1","pages":"333-340"},"PeriodicalIF":0.0,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85018853","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137749
J. Navarro, R. Evard, Daniel Nurmi, N. Desai
Systems administrators of large clusters often need to perform the same administrative task hundreds or thousands of times. Administrators have traditionally performed some time-consuming tasks, such as operating system installation, configuration, and maintenance, manually. By combining network services such as DHCP, TFTP, FTP, HTTP, and NFS with remote hardware control and scripted installation, configuration, and maintenance techniques, cluster administrators can automate these administrative tasks. Scalable cluster administration addresses this challenge: What hardware and software design techniques can cluster builders use to automate cluster administration on very large clusters? We describe the approach used in the Mathematics and Computer Science Division of Argonne National Laboratory on Chiba City I, a 314-node Linux cluster; and we analyze the scalability, flexibility, performance and reliability benefits and limitations from that approach.
{"title":"Scalable cluster administration - Chiba City I approach and lessons learned","authors":"J. Navarro, R. Evard, Daniel Nurmi, N. Desai","doi":"10.1109/CLUSTR.2002.1137749","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137749","url":null,"abstract":"Systems administrators of large clusters often need to perform the same administrative task hundreds or thousands of times. Administrators have traditionally performed some time-consuming tasks, such as operating system installation, configuration, and maintenance, manually. By combining network services such as DHCP, TFTP, FTP, HTTP, and NFS with remote hardware control and scripted installation, configuration, and maintenance techniques, cluster administrators can automate these administrative tasks. Scalable cluster administration addresses this challenge: What hardware and software design techniques can cluster builders use to automate cluster administration on very large clusters? We describe the approach used in the Mathematics and Computer Science Division of Argonne National Laboratory on Chiba City I, a 314-node Linux cluster; and we analyze the scalability, flexibility, performance and reliability benefits and limitations from that approach.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":"95 1","pages":"215-221"},"PeriodicalIF":0.0,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76857516","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137755
Ming Li, Wenchao Tao, Dani Goldberg, I. Hsu, Y. Tamir
We describe the communication infrastructure (CI) for our fault-tolerant cluster middleware, which is optimized for two classes of communication: for the applications and for the cluster management middleware. This CI was designed for portability and for efficient operation on top of modern user-level message passing mechanisms. We present a functional fault model for the CI and show how platform-specific faults map to this fault model. Based on this fault model, we have developed a fault injection scheme that is integrated with the CI and is thus portable across different communication technologies. We have used fault injection to validate and evaluate the implementation of the CI itself as well as the cluster management middleware in the presence of communication faults.
{"title":"Design and validation of portable communication infrastructure for fault-tolerant cluster middleware","authors":"Ming Li, Wenchao Tao, Dani Goldberg, I. Hsu, Y. Tamir","doi":"10.1109/CLUSTR.2002.1137755","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137755","url":null,"abstract":"We describe the communication infrastructure (CI) for our fault-tolerant cluster middleware, which is optimized for two classes of communication: for the applications and for the cluster management middleware. This CI was designed for portability and for efficient operation on top of modern user-level message passing mechanisms. We present a functional fault model for the CI and show how platform-specific faults map to this fault model. Based on this fault model, we have developed a fault injection scheme that is integrated with the CI and is thus portable across different communication technologies. We have used fault injection to validate and evaluate the implementation of the CI itself as well as the cluster management middleware in the presence of communication faults.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":"191 1","pages":"266-274"},"PeriodicalIF":0.0,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76930907","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137748
K. Pedretti, R. Brightwell, Joshua Williams
In this paper, we describe additions and modifications to the Computational Plant (Cplant/sup /spl trade//) system software to support multi-processor compute nodes and to support heterogeneous node types. We describe how these capabilities have been incorporated into our scalable runtime system and how these changes affect the interface seen by end users and application developers. We also discuss several important operating system and networking issues that can directly impact application performance. We present some initial performance metrics that indicate how our current implementation scales when multiple processes are running on a single node.
{"title":"Cplant/sup /spl trade// runtime system support for multi-processor and heterogeneous compute nodes","authors":"K. Pedretti, R. Brightwell, Joshua Williams","doi":"10.1109/CLUSTR.2002.1137748","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137748","url":null,"abstract":"In this paper, we describe additions and modifications to the Computational Plant (Cplant/sup /spl trade//) system software to support multi-processor compute nodes and to support heterogeneous node types. We describe how these capabilities have been incorporated into our scalable runtime system and how these changes affect the interface seen by end users and application developers. We also discuss several important operating system and networking issues that can directly impact application performance. We present some initial performance metrics that indicate how our current implementation scales when multiple processes are running on a single node.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":"39 1","pages":"207-214"},"PeriodicalIF":0.0,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82950919","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137731
Qianfeng Zhang, C. Keppitiyagama, Alan S. Wagner
We present work that extends our previous Myrinet port for LAM/MPI, MPI-NP, with collective communication primitives on the NIC. This work is another step in our experiment of making the NIC MPI aware. We believe that an MPI aware control program on the NIC can deliver a richer set of performance enhancements, not just restricted to better bandwidth/latency, to MPI applications. MPI collective communication involves considerable interactions between the communication subsystems of the nodes that are not of any direct interest to the application. By migrating these talkative components to the Myrinet network interface card we allow this dialog between the nodes to happen with minimum latency. We explore the advantage of supporting several MPI collective communication routines on the NIC. These include MPI /spl I.bar/Bcast (), MPI/spl I.bar/Barrier and MPI/spl I.bar/Comm/spl I.bar/Create ().
{"title":"Supporting MPI collective communication on network processors","authors":"Qianfeng Zhang, C. Keppitiyagama, Alan S. Wagner","doi":"10.1109/CLUSTR.2002.1137731","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137731","url":null,"abstract":"We present work that extends our previous Myrinet port for LAM/MPI, MPI-NP, with collective communication primitives on the NIC. This work is another step in our experiment of making the NIC MPI aware. We believe that an MPI aware control program on the NIC can deliver a richer set of performance enhancements, not just restricted to better bandwidth/latency, to MPI applications. MPI collective communication involves considerable interactions between the communication subsystems of the nodes that are not of any direct interest to the application. By migrating these talkative components to the Myrinet network interface card we allow this dialog between the nodes to happen with minimum latency. We explore the advantage of supporting several MPI collective communication routines on the NIC. These include MPI /spl I.bar/Bcast (), MPI/spl I.bar/Barrier and MPI/spl I.bar/Comm/spl I.bar/Create ().","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":"16 1","pages":"75-82"},"PeriodicalIF":0.0,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87070219","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137771
M. Janjua, M. M. Yasin, Ch. Falak Sher, K. Awan, I. Hassan
CEJVM is a cluster enabled Java Virtual Machine, which executes in a distributed fashion among collaborating nodes of a dedicated cluster. It extends Java's multithreading mechanism to the parallel computing paradigm by transparent migration of independent application threads modeled in master worker paradigm. The goal is to obtain improved performance for computationally-intensive multi-threaded Java programs without modifying traditional JVM code, Java language or compiler implementation. Deploying a master worker relationship among the nodes in the cluster, CEJVM monitors, packs, transports and resurrects the Java threads on "Master JVM" and helper "Worker JVMs". Currently, we have created a prototype that runs pure Java applications on the local Ethernet based cluster of Win2K computers. We have achieved appreciable speedup for an ANN training program written in Java (with independent threads) on CEJVM without any modification to the program's source code.
{"title":"CEJVM: \"cluster enabled Java Virtual Machine\"","authors":"M. Janjua, M. M. Yasin, Ch. Falak Sher, K. Awan, I. Hassan","doi":"10.1109/CLUSTR.2002.1137771","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137771","url":null,"abstract":"CEJVM is a cluster enabled Java Virtual Machine, which executes in a distributed fashion among collaborating nodes of a dedicated cluster. It extends Java's multithreading mechanism to the parallel computing paradigm by transparent migration of independent application threads modeled in master worker paradigm. The goal is to obtain improved performance for computationally-intensive multi-threaded Java programs without modifying traditional JVM code, Java language or compiler implementation. Deploying a master worker relationship among the nodes in the cluster, CEJVM monitors, packs, transports and resurrects the Java threads on \"Master JVM\" and helper \"Worker JVMs\". Currently, we have created a prototype that runs pure Java applications on the local Ethernet based cluster of Win2K computers. We have achieved appreciable speedup for an ANN training program written in Java (with independent threads) on CEJVM without any modification to the program's source code.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":"62 3 1","pages":"389-393"},"PeriodicalIF":0.0,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87738134","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137732
Rinku Gupta, V. Tipparaju, J. Nieplocha, D. Panda
Most high performance scientific applications require efficient support for collective communication. Point-to-point message-passing communication in current generation clusters are based on the Send/Recv communication model. Collective communication operations built on top of such point-to-point message-passing operations might achieve suboptimal performance. VIA and the emerging InfiniBand architecture support remote DMA operations, which allow data to be moved between the nodes with low overhead; they also allow to create and provide a logical shared memory address space across the nodes. In this paper we focus on barrier, a frequently-used collective operations. We demonstrate how RDMA write operations can be used to support an inter-node barrier in a cluster with SMP nodes. Combining this with a scheme to exploit shared memory within a SMP node, we develop a fast barrier algorithm for a cluster of SMP nodes with a cLAN VIA interconnect. Compared to current barrier algorithms using the Send/Recv communication model, the new approach is shown to reduce barrier latency on a 64 processor (32 dual nodes) system by up to 66%. These results demonstrate that high performance and scalable barrier implementations can be delivered on current and next generation VIA/Infiniband-based clusters with RDMA support.
{"title":"Efficient barrier using remote memory operations on VIA-based clusters","authors":"Rinku Gupta, V. Tipparaju, J. Nieplocha, D. Panda","doi":"10.1109/CLUSTR.2002.1137732","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137732","url":null,"abstract":"Most high performance scientific applications require efficient support for collective communication. Point-to-point message-passing communication in current generation clusters are based on the Send/Recv communication model. Collective communication operations built on top of such point-to-point message-passing operations might achieve suboptimal performance. VIA and the emerging InfiniBand architecture support remote DMA operations, which allow data to be moved between the nodes with low overhead; they also allow to create and provide a logical shared memory address space across the nodes. In this paper we focus on barrier, a frequently-used collective operations. We demonstrate how RDMA write operations can be used to support an inter-node barrier in a cluster with SMP nodes. Combining this with a scheme to exploit shared memory within a SMP node, we develop a fast barrier algorithm for a cluster of SMP nodes with a cLAN VIA interconnect. Compared to current barrier algorithms using the Send/Recv communication model, the new approach is shown to reduce barrier latency on a 64 processor (32 dual nodes) system by up to 66%. These results demonstrate that high performance and scalable barrier implementations can be delivered on current and next generation VIA/Infiniband-based clusters with RDMA support.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":"131 1","pages":"83-90"},"PeriodicalIF":0.0,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74901901","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137741
Yu Chen, Xiaoge Wang, Z. Jiao, Jun Xie, Zhihui Du, Sanli Li
Virtual Interface Architecture (VIA) established a communication model with low latency and high bandwidth, and defined the standard of user-level high-performance communication specification in cluster systems. This paper analyzes the current development, principle and implementations of VIA, and presents user-level high-performance communication software, MyVIA, based on Myrinet, which is comfortable with VIA specification. The paper first describes the design principle and framework of MyVIA, then proposes new technologies of MyVIA including User TLB, continued host physical memory and varied NIC buffer, the pipelining communication based on resource and DMA chain, and physical descriptor ring. Experimental results of performance comparisons and analysis are presented; the one-way bandwidth of MyVIA for a 4 KB message is 250 MB/s, and the lowest one-way latency is 8.46 /spl mu/s, which shows that the performance of MyVIA surpassed that of other implementations of VIA.
{"title":"MyVIA: a design and implementation of the high performance Virtual Interface Architecture","authors":"Yu Chen, Xiaoge Wang, Z. Jiao, Jun Xie, Zhihui Du, Sanli Li","doi":"10.1109/CLUSTR.2002.1137741","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137741","url":null,"abstract":"Virtual Interface Architecture (VIA) established a communication model with low latency and high bandwidth, and defined the standard of user-level high-performance communication specification in cluster systems. This paper analyzes the current development, principle and implementations of VIA, and presents user-level high-performance communication software, MyVIA, based on Myrinet, which is comfortable with VIA specification. The paper first describes the design principle and framework of MyVIA, then proposes new technologies of MyVIA including User TLB, continued host physical memory and varied NIC buffer, the pipelining communication based on resource and DMA chain, and physical descriptor ring. Experimental results of performance comparisons and analysis are presented; the one-way bandwidth of MyVIA for a 4 KB message is 250 MB/s, and the lowest one-way latency is 8.46 /spl mu/s, which shows that the performance of MyVIA surpassed that of other implementations of VIA.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":"511 1","pages":"160-167"},"PeriodicalIF":0.0,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83789813","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137790
M. Pourzandi, I. Haddad, C. Levert, Miroslaw Zakrzewski, M. Dagenais
Traditionally the telecom industry has used clusters to meet its carrier-class requirements of high availability, reliability, and scalability, while relying on cost-effective hardware and software. Efficient cluster security is now an essential requirement and has not yet been addressed in a coherent fashion on clustered systems. This paper presents an approach for distributed security architecture that supports advanced security mechanisms for current and future security needs, targeted for carrier-class application servers running on clustered systems.
{"title":"A new architecture for secure carrier-class clusters","authors":"M. Pourzandi, I. Haddad, C. Levert, Miroslaw Zakrzewski, M. Dagenais","doi":"10.1109/CLUSTR.2002.1137790","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137790","url":null,"abstract":"Traditionally the telecom industry has used clusters to meet its carrier-class requirements of high availability, reliability, and scalability, while relying on cost-effective hardware and software. Efficient cluster security is now an essential requirement and has not yet been addressed in a coherent fashion on clustered systems. This paper presents an approach for distributed security architecture that supports advanced security mechanisms for current and future security needs, targeted for carrier-class application servers running on clustered systems.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":"31 1","pages":"494-497"},"PeriodicalIF":0.0,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83879695","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137751
E. Cecchet
Distributed Shared Memories (DSM) performance has always suffered from high network latencies and software communication layers with a large overhead. Memory mapped networks such as Scalable Coherent Interface (SCI) allow to reliably access remote memory without involving the operating system. To show how DSM systems can benefit from this technology, we have developed SciFS, a DSM tightly integrated with the operating system, that exploits the high performance and the remote memory access capabilities of SCI. We first show the respective advantages of two communications techniques with SCI: programmed IO (PIO) and remote DMA (RDMA). Then, we describe how to build a scalable page transfer mechanism by mixing PIO and RDMA. Despite the lack of a broadcast mechanism with SCI, we demonstrate that it is possible to build scalable synchronization primitives using PIO. Finally, we evaluate various consistency models with scientific computing applications from the Splash benchmark. We observe that, even if the rough network performance is good, it is not sufficient to obtain acceptable results with applications that require fine grain parallelism. However, we show that memory mapped networks provide an efficient hardware support to implement software DSM systems without requiring complex relaxed consistency models. This way, DSM design can be greatly simplified using this technology.
{"title":"Memory mapped networks: a new deal for distributed shared memories ? the SciFS experience","authors":"E. Cecchet","doi":"10.1109/CLUSTR.2002.1137751","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137751","url":null,"abstract":"Distributed Shared Memories (DSM) performance has always suffered from high network latencies and software communication layers with a large overhead. Memory mapped networks such as Scalable Coherent Interface (SCI) allow to reliably access remote memory without involving the operating system. To show how DSM systems can benefit from this technology, we have developed SciFS, a DSM tightly integrated with the operating system, that exploits the high performance and the remote memory access capabilities of SCI. We first show the respective advantages of two communications techniques with SCI: programmed IO (PIO) and remote DMA (RDMA). Then, we describe how to build a scalable page transfer mechanism by mixing PIO and RDMA. Despite the lack of a broadcast mechanism with SCI, we demonstrate that it is possible to build scalable synchronization primitives using PIO. Finally, we evaluate various consistency models with scientific computing applications from the Splash benchmark. We observe that, even if the rough network performance is good, it is not sufficient to obtain acceptable results with applications that require fine grain parallelism. However, we show that memory mapped networks provide an efficient hardware support to implement software DSM systems without requiring complex relaxed consistency models. This way, DSM design can be greatly simplified using this technology.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":"97 1","pages":"231-238"},"PeriodicalIF":0.0,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85266469","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：