S. Englert, J. Gray, T. Kocher, and P. Shah. A benchmark of NonStop SQL release 2 demonstrating nearlinear speedup and scaleup on large databases. In SIGMETRICS90, pages 245-246, May 1990.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....performance. Our study indicates that the sampling technique is the better approach. 1 Introduction There are several architectures for designing multiprocessor database computers. However, the multicomputer model [14] is most popularly used for its scalability to support very large databases [1, 2, 4, 7, 12, 17, 18]. In this architecture, the processing nodes (PNs) are interconnected through a communication network. Each PN has its own private memory and dedicated disk drives. Parallel database systems implemented for this hardware structure are popularly known as shared nothing (SN) systems [16] The join ....

S. Englert, J. Gray, T. Kocher, and P. Shah. A benchmark of NonStop SQL release 2 demonstrating nearlinear speedup and scaleup on large databases. In SIGMETRICS90, pages 245-246, May 1990.

....River solves the additional problem of ensuring that these applications perform robustly even when underlying hardware and software components perform erratically. Although previous work has addressed how to design large scale systems that can tolerate correctness faults in individual components [7, 17, 32, 34], little work has focused on how to design systems that tolerate performance faults, where a performance fault is the unexpected low performance of a component within the system. Within clusters, performance faults are commonplace [2] especially for modern disks. Some of the reasons static and ....

S. Englert, J. Gray, T. Kocher, and P. Shah. A benchmark of NonStop SQL Release 2 demonstrating near-linear speedup and scaleup on large databases. In Proceedings of the

....concurrency achievable determines the degree of parallelism can be attained. Several parallel computer architectures are suitable for database applications. Among them, the multicomputer architecture [18] has been recognized as having the most scalable structure for handling very large databases [2, 3, 6, 10, 15, 21, 22]. In this architecture, the processing nodes (PNs) are interconnected through a communication network. Each PN has its own private memory and dedicated disk drives. In order to take advantage of such a machine organization, the database management system must be carefully designed to overcome the ....

Susanne Englert, Jim Gray, Terry Kocher, and Praful Shah. A benchmark of NonStop SQL release 2 demonstrating near-linear speedup and scaleup on large databases. In SIGMETRICS90, pages 245--246, May 1990.

....manage the data. Tuples within a relation are partitioned across the AMPs so that the DBC 1012 can support parallelism both within and between independent requests. The AMPs and IFPs are interconnected by a proprietary Ynet active logic interconnection network. The Tandem NonStop SQL system [27, 28, 33, 16] is essentially a distributed relational database system. Data objects are partitioned across multiple processing nodes and transactions can access data at different sites. The system provides local autonomy so that a site can perform work despite failures at other sites or in the interconnection ....

Susanne Englert, Jim Gray, Terrye Kocher, and Praful Shah. A Benchmark of NonStop SQL Release 2 Demonstrating Near-Linear Speedup and Scaleup on Large Databases. Technical Report 89.4, Tandem Computers Inc., May 1989.

....as the workload is increased by adding a proportional number of processors and disks. Speedup is an interesting metric because it indicates whether additional processors and disks result in a corresponding decrease in the response time of a query. A similar set of experiments were reported in [EGKS89] for equi join queries on Release 2 of Tandem s NonStop SQL system, in [DGS 90] for equi join queries in Gamma, and in [DNS91] for non equijoin queries in Gamma. Scaleup and speedup results for several parallel sorting algorithms are contained in [Gra90] and in [STG 90] A third ....

S. Englert, J. Gray, T. Kocher, and P. Shah. A benchmark of Nonstop SQL Release 2 demonstrating near-linear speedup and scaleup on large database. Technical Report 89.4, Tandem Part No. 27469, Tandem Computers, 1989.

....hardware. None of these papers compared the hashing join algorithms with nested loop with index algorithms. Valduriez and Gardarin [VG84] compared parallel join and semijoin algorithms based on hashing, sort merge, and nested loops, but did not consider nested loops with index. The Tandem group [EGKS90] state that their NonStop SQL system uses a variant of parallel nested loops with index algorithm if the appropriate indices exist and one of the relations is small, and hashing followed by sort merge otherwise, but did not compare the two algorithms. Wolf et al. WDYT90, WDY90] consider the ....

....originally proposed in [ESW78] with the fragment phase a no op since this algorithm begins with S fragmented about the sites of the system. RNL is also the algorithm used by Tandem if the appropriate index exists, one of the relations is small, and the join is on a key for the small relation [EGKS90] Recently Stamos and Young have proposed an improvement on the full fragment replicate algorithm [SY89] that partially replicates and redistributes both input relations. While this algorithm improves on the network cost of a full fragment replicate, it is not appropriate for use in RNL because ....

S. Englert, J. Gray, T. Kocher, and P. Shah. A benchmark of NonStop SQL Release 2 demonstrating near-linear speedup and scaleup on large database. In Proc. of the SIGMETRICS Conference, pages 245--247, May 1990.

....ffl shared nothing (SN) architectures: parallel systems, in which each processor is only able to access directly its local memory and its disks; data exchange through a network allows to access to memories and disks of other processors. Such architectures are for instance Tandem NonStop SQL [30, 6] and Teradata s DBC 1012 [20] Research partially supported by DFG, SFB 124 and Siemens AG. ffl shared everything (SE) architectures: every processor has direct access to a global memory and to global disks. So far they have been built only with a small number of processors. Examples are the ....

Englert, S., Gray, J., Kocher, T., Shath, P.: A Benchmark of NonStop SQL Release 2 Demonstrating Near--Linear Speedup and Scale--Up on Large Databases. Proc. ACM SIGMETRICS Conf., (1990)

....not all, of the above parallel file systems are designed for machines with an architecture of this type. There are several database machines that can filter tuples from blocks of data as they are read off disk, forwarding only those of interest on to the computational nodes (e.g. Tandem NonStop [EGKS90] The Super Database Computer [KHH 92] has disk controllers that continuously produce tasks from the input data set, which are consumed and processed by CPs as they become available. Thus they have a load dependent data distribution mechanism. The Bridge (PIFS) file system [DSE88] tried to ....

Susanne Englert, Jim Gray, Terrye Kocher, and Praful Shah. A benchmark of NonStop SQL Release 2 demonstrating near-linear speedup and scaleup on large databases. In Proceedings of the 1990 ACM Sigmetrics Conference on Measurement and Modeling of Computer Systems, pages 245--246, May 1990.

....disk directed I O focuses more on when data flows (for better disk and cache performance) Some parallel database machines use an architecture similar to disk directed I O, in that certain operations are moved closer to the disks to allow for more optimization. In the Tandem NonStop system [EGKS90] each query is sent to all IOPs, which scan the local database partition and send only the relevant tuples back to the requesting node. The Super Database Computer [KHH 92] has disk controllers that continuously produce tasks from the input data set, which are consumed and processed by CPs as ....

Susanne Englert, Jim Gray, Terrye Kocher, and Praful Shah. A benchmark of NonStop SQL Release 2 demonstrating near-linear speedup and scaleup on large databases. In Proceedings of the 1990 ACM Sigmetrics Conference on Measurement and Modeling of Computer Systems, pages 245--246, May 1990.

....disk directed I O focuses more on when data flows (for better disk and cache performance) Some parallel database machines use an architecture similar to disk directed I O, in that certain operations are moved closer to the disks to allow for more optimization. In the Tandem NonStop system [EGKS90] each query is sent to all IOPs, which scan the local database partition and send only the relevant tuples back to the requesting node. The Super Database Computer [KHH 92] has disk controllers that continuously produce tasks from the input data set, which are consumed and processed by CPs as ....

Susanne Englert, Jim Gray, Terrye Kocher, and Praful Shah. A benchmark of NonStop SQL Release 2 demonstrating near-linear speedup and scaleup on large databases. In Proceedings of the 1990 ACM Sigmetrics Conference on Measurement and Modeling of Computer Systems, pages 245--246, May 1990.

....to implement a distributed database on a configuration that consists of a moderate to large number of small, inexpensive processors that do not share any storage and thus communicate by sending messages. Several of these distributed database systems have been designed, prototyped, or even marketed [4, 7, 11, 16, 19]. Such multiprocessor systems can exploit inter query parallelism for simple transactions and intra query parallelism for long, complex queries by spreading data and processing across all processors in the system. A transaction that manipulates data stored on several processors is termed ....

S. Englert, J. Gray, T. Kocher, and P. Shah. A benchmark of NonStop SQL release 2 demonstrating near-linear speedup and scaleup on large databases. Technical Report 89.4, Tandem Computer Inc., May 1989.

....disk directed I O focuses more on when data flows (for better disk and cache performance) Some parallel database machines use an architecture similar to disk directed I O, in that certain operations are moved closer to the disks to allow for more optimization. In the Tandem NonStop system [EGKS90] each query is sent to all IOPs, which scan the local database partition and send only the relevant tuples back to the requesting node. The Super Database Computer [KHH 92] has disk controllers that continuously produce tasks from the input data set, which are consumed and processed by CPs as ....

Susanne Englert, Jim Gray, Terrye Kocher, and Praful Shah. A benchmark of NonStop SQL Release 2 demonstrating near-linear speedup and scaleup on large databases. In Proceedings of the 1990 ACM Sigmetrics Conference on Measurement and Modeling of Computer Systems, pages 245--246, May 1990.

....environments, not only must such systems perform correctly, but they must also operate with high performance. Much of the previous work in distributed computing has addressed the design of large scale systems that function correctly, in spite of correctness faults of individual components [18, 49, 82, 86]. However, there has been little development of techniques to tolerate performance faults unexpected performance fluctuations from the components that comprise the system. Due to this shortcoming, many systems are overly sensitive to performance variations, in that global performance is high if ....

....account how global performance characteristics will be altered under component performance variations. Fortunately, much of the previous work in the field of distributed computing has addressed the design of large scale systems that can tolerate such correctness faults in individual components [18, 24, 49, 61, 82, 86, 107, 114]. The practical notion behind such work is that distributed and parallel systems consist of both hardware and software components that will periodically fail; a system that works continuously on top of such unreliable components must be designed to operate in spite of such failures. A good example ....

Susanne Englert, Jim Gray, Terrye Kocher, and Praful Shah. A benchmark of NonStop SQL Release 2 demonstrating near-linear speedup and scaleup on large databases. In Proceedings of the 1990 ACM SIGMETRICS Conference on Measurement and Modeling of Computer Systems, pages 245--246, May 1990.

No context found.

EGKS90 Englert, S., Gray, J., Kocher, T., Shath, P. 1990: A Benchmark of NonStop SQL Release 2 Demonstrating Near-Linear Speedup and Scale-Up on Large Databases. Proc. ACM SIGMETRICS Conf., 245-246.

No context found.

EGKS90 Englert, S., Gray, J., Kocher, T., Shath, P.: A Benchmark of NonStop SQL Release 2 Demonstrating Near-Linear Speedup and Scale-Up on Large Databases. Proc. ACM SIGMETRICS Conf., 245-246 (1990).

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC