Tradeoffs Between Coupling Small and Large Processors for Transaction Processing

A methodology is developed to determine the number of processors needed to satisfy transaction throughput and response time requirements for processors of different MIPS (sizes). The minimum MIPS per processor required to satisfy response time and throughput constraints in a transaction processing complex of N coupled systems is also determined. For realistic overhead assumptions, despite large assumed cost advantages on a per-MIPS basis, it is found that very small systems may not match up to the cost/performance of some larger systems, when required to meet the same throughput and response-time constraints. If transactions running on smaller systems were allowed a larger response-time constraint, then it may be possible to construct a lower-cost system from smaller and less expensive processors, generally with lower supportable maximum throughput. Besides the coupling degradation between multiprocessor systems, there is a small systems effect. The cost criterion indicates that there is an optimum processor size below which total system costs would increase appreciably. >

[1]  Philip S. Yu,et al.  On Affinity Based Routing in Multi-System Data Sharing , 1986, VLDB.

[2]  Philip S. Yu,et al.  On multisystem coupling through function request shipping , 1986, IEEE Transactions on Software Engineering.

[3]  David K. Gifford,et al.  The TWA reservation system , 1984, CACM.

[4]  Peter P. Uhrowczik,et al.  IMS/VS: An Evolving System , 1982, IBM Syst. J..

[5]  Philip S. Yu,et al.  Analysis of Fault Tolerant Multiprocessor Architectures for Lock Engine Design , 1987, Computer systems science and engineering.

[6]  Stephen S. Lavenberg A simple analysis of exclusive and shared lock contention in a database system , 1984, SIGMETRICS '84.

[7]  Philip S. Yu,et al.  Analysis of Affinity Based Routing in Multi-System Data Sharing , 1987, Perform. Evaluation.

[8]  Elliot Nestle,et al.  The SYNAPSE N+1 System: architectural characteristics and performance data of a tightly-coupled multiprocessor system , 1985, ISCA '85.

[9]  Joel F. Bartlett,et al.  A NonStop kernel , 1981, SOSP.

[10]  Philip S. Yu,et al.  Modelling of centralized concurrency control in a multi-system environment , 1985, SIGMETRICS '85.

[11]  Philip S. Yu,et al.  Performability Analysis of Operation Modes of Configurable Duplex Systems , 1986, FJCC.

[12]  Nancy P. Kronenberg,et al.  VAXcluster: a closely-coupled distributed system , 1986, TOCS.

[13]  Philip M. Neches,et al.  The Genesis of a Database Computer , 1984, Computer.

[14]  Peter B. Mark The Sequoia computer: a fault-tolerant tightly-coupled multiprocessor architecture , 1985, ISCA 1985.

[15]  Armond Inselberg,et al.  Synapse tightly coupled multiprocessors: a new approach to solve old problems , 1984, AFIPS '84.

[16]  Philip S. Yu,et al.  Design and Analysis of Integrated Concurrency-Coherence Controls , 1987, VLDB.

[17]  J. T. Robinson,et al.  On coupling multi-systems through data sharing , 1987, Proceedings of the IEEE.

[18]  Kazuo Goto,et al.  The DCS: a new approach to multisystem data-sharing , 1984, AFIPS '84.

[19]  Philip S. Yu,et al.  A hybrid data sharing-data partitioning architecture for transaction processing , 1988, Proceedings. Fourth International Conference on Data Engineering.

[20]  Philip S. Yu,et al.  On Coupling Partitioned Database Systems , 1986, ICDCS.

[21]  Gordon Bell,et al.  The Mini and Micro Industries , 1984, Computer.

[22]  Phillip Ein-Dor Grosch's law re-revisited: CPU power and the cost of computation , 1985, CACM.

[23]  Robert W. Horst,et al.  An architecture for high volume transaction processing , 1985, ISCA '85.