Quantitative system performance: computer system analysis using queueing network models
Quantitative system performance: computer system analysis using queueing network models
Multi-disk management algorithms
SIGMETRICS '87 Proceedings of the 1987 ACM SIGMETRICS conference on Measurement and modeling of computer systems
Access path selection in a relational database management system
SIGMOD '79 Proceedings of the 1979 ACM SIGMOD international conference on Management of data
Proceedings of the 2nd International Workshop on High Performance Transaction Systems
GAMMA - A High Performance Dataflow Database Machine
VLDB '86 Proceedings of the 12th International Conference on Very Large Data Bases
Process and dataflow control in distributed data-intensive systems
SIGMOD '88 Proceedings of the 1988 ACM SIGMOD international conference on Management of data
SIGMOD '88 Proceedings of the 1988 ACM SIGMOD international conference on Management of data
DPDS '88 Proceedings of the first international symposium on Databases in parallel and distributed systems
Using CSIM to model complex systems
WSC '88 Proceedings of the 20th conference on Winter simulation
Prototyping Bubba, A Highly Parallel Database System
IEEE Transactions on Knowledge and Data Engineering
| Hi-index | 0.00 |
In dataflow architectures, each dataflow node (i.e., operation) is typically executed on a single physical node. We are concerned with distributed data-intensive systems, in which each base (i.e., persistent) set of data has been declustered over many physical nodes to achieve load balancing. Because of large base set size, each operation is executed where the base set resides, and intermediate results are transferred between physical nodes. In such systems, each dataflow node is typically executed on many physical nodes. Furthermore, because computations are data-dependent, we cannot know until run time which subset of the physical nodes containing a particular base set will be involved in a given dataflow node. This uncertainty affects program loading, task activation and termination, and data transfer among the nodes.In this paper we focus on the problem of how a dataflow node in such an environment knows when it has received data from all the physical nodes from which it is ever going to receive. We call this the dataflow control problem. The interesting part of the problem is trying to achieve correctness efficiently. We propose three solutions to this problem, and compare them quantitatively by the metrics of total message traffic, message system throughput and data transfer response time.