Selecting buckets for redistributing data between nodes in a parallel database in the incremental mode

• US 5,687,369 A
• Filed: 09/02/1993
• Issued: 11/11/1997
• Est. Priority Date: 09/02/1993
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Li, Shih-Gong
• Primary Examiner(s): Black, Thomas G.
• Assistant Examiner(s): Von Buhr, Maria N.
• Application Number:
  US08/116,089
• Time in Patent Office: 1,531 Days
• Field of Search: 364/DIG. 1, 364/DIG. 2, 395/200, 395/600, 395/775, 395/619
• US Class Current: 1/1
• CPC Class Codes: G06F 16/27 Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
• CPC Class Codes: Y10S 707/99954 Version management

Abstract

A method, system and program for determining a set of tables from which data can be redistributed between nodes in a parallel database. First, a list of candidate tables from those stored in the memories associated with a plurality of nodes in the parallel database is established. The list is ordered according a workload of each of the candidate tables. Next, each of the candidate tables is tested to determine whether it can be scanned at each of the plurality of nodes at which the candidate table is stored within a predetermined time period. Any candidate table which can be scanned within the predetermined time period is added to the set of tables from which data can be redistributed. Once the set of tables for redistribution are established, data sets, or buckets, within the redistribution tables are selected. The selected buckets are redistributed from the plurality of nodes to a receiving node. The receiving node is a new node to the parallel database system or an underutilized node in the parallel database system.

Claims

1. A process for redistributing data between nodes in a parallel database system, wherein the redistributing process is operating concurrently with other processes, each process operating in a respective time period allotted to the respective process, comprising the steps of:

determining redistributable data from at least one transmitting node which may be redistributed to a receiving node within a redistribution time period allotted to the redistributing process; and

redistributing the redistributable data from the at least one transmitting node to the receiving node within the redistribution time period.

2. The method as recited in claim 1 wherein the determining step further comprises the steps of:

establishing a list of candidate tables stored in the at least one transmitting node, the list ordered according a workload of each of the candidate tables;

determining whether each of the candidate tables can be scanned within the at least one transmitting node at which the candidate table is stored within the redistribution time period; and

adding any candidate table which can be scanned within the redistribution time period to a set of tables from which the redistributable data is determined.

3. The method as recited in claim 2 further comprising the step of selecting the redistributable data from the set of tables.

4. The method as recited in claim 2 wherein the receiving node is a new node to the parallel database system.

5. The method as recited in claim 2 wherein the receiving node is an underutilized node in the parallel database system.

6. A system for redistributing data between nodes in a parallel database system in which a redistribution process operates concurrently with other processes, each process operating in a respective time period allotted to the respective process, comprising:

means for determining redistributable data from at least one transmitting node which may be redistributed to a receiving node within a redistribution time period allotted to the redistribution process; and

means for redistributing the redistributable data from the at least one transmitting node to the receiving node within the redistribution time period.

7. The system as recited in claim 6 wherein the determining means further comprises:

means for establishing a list of candidate tables stored in memories associated with a plurality of transmitting nodes in the parallel database system, the list ordered according to a workload of each of the candidate tables;

means for determining whether each of the candidate tables can be scanned at each of the plurality of nodes at which the candidate table is stored within the redistribution and

means for adding any candidate table which can be scanned within the redistribution time period to a set of tables from which the redistributable data is determined.

8. The system as recited in claim 7 further comprising means for selecting the redistributable data from the set of tables.

9. The system as recited in claim 7 wherein the receiving node is a new node to the parallel database system.

10. The system as recited in claim 7 wherein the receiving node is an underutilized node in the parallel database system.

11. A computer program product readable in a computer memory for redistributing data between nodes in a parallel database system, which when executed causes a redistribution process to operate concurrently with other processes, each process operating in a respective time period allotted to the respective process, the product comprising:

means for determining redistributable data from at least one transmitting node which may be redistributed to a receiving node within a redistribution time period allotted to the redistribution process; and

means for redistributing the redistributable data from the at least one transmitting node to the receiving node within the redistribution time period.

12. The product as recited in claim 11 wherein the determining means further comprises:

means for establishing a list of candidate tables stored in memories associated with a plurality of transmitting nodes in the parallel database system, the list ordered according to a workload of each of the candidate tables;

means for determining whether each of the candidate tables can be scanned at each of the plurality of nodes at which the candidate table is stored within the redistribution time period; and

means for adding any candidate table which can be scanned within the redistribution time period to a set of tables from which the redistributable data is determined.

13. The product as recited in claim 12 further comprising means for selecting the redistributable data the set of tables.

14. The product as recited in claim 12 wherein the receiving node is a new node to the parallel database system.

15. The product as recited in claim 12 wherein the receiving node is an underutilized node in the parallel database system.