Method and apparatus for performing consistent reads in multiple-server environments

• US 5,832,521 A
• Filed: 02/28/1997
• Issued: 11/03/1998
• Est. Priority Date: 02/28/1997
• Status: Expired due to Term

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle Corporation
• Inventors: Klots, Boris, Bamford, Roger J.
• Primary Examiner(s): Black, Thomas G.
• Assistant Examiner(s): Mizrahi, Diane D.
• Application Number:
  US08/808,622
• Time in Patent Office: 613 Days
• Field of Search: 1/1, 707/4, 707/10, 707/103, 379/111, 380/25, 382/125, 382/155, 395/200.33, 395/200.47, 395/200.49, 395/683
• US Class Current: 709/219
• 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: G06F 16/275 Synchronous replication
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99953 Recoverability
• CPC Class Codes: Y10S 707/99954 Version management

Abstract

A method and apparatus for supplying a particular version of a data item to a transaction executing in a first database server is provided. The data item is in a database accessible by the first database server and one or more other database servers. The method involves determining whether a current version of the data item resides in a cache that is remote relative to the first database server. If the current version of the data item does not reside in a cache that is remote relative to the first database server, then a first derivation mechanism that is local to the first database server to derives the particular version of the data item from the current version of the data item. If the current version of the data item resides in a cache that is remote relative to the first database server, then a second derivation mechanism that is local to the cache derives the particular version of the data item from the current version of the data item. Once built, the particular version of the data item sent to the first database server. Determining whether a current version of the data item resides in a cache that is remote relative to the first database server may be performed by determining whether a process currently holds a write lock on the data item, and if a process currently holds a write lock on the data item, then determining whether the process is remote relative to the first database server.

Claims

1. A method for supplying a particular version of a data item to a transaction executing in a first database server, where the data item is in a database accessible by the first database server and one or more other database servers, the method comprising the steps of:

determining whether a current version of the data item resides in a cache that is remote relative to the first database server;

if the current version of the data item does not reside in a cache that is remote relative to the first database server, then causing a first derivation mechanism that is local to the first database server to derive the particular version of the data item from the current version of the data item; and

if the current version of the data item resides in a cache that is remote relative to the first database server, then

causing a second derivation mechanism that is local to the cache to derive the particular version of the data item from the current version of the data item; and

sending the particular version of the data item to the first database server.

2. The method of claim 1 wherein the step of determining whether a current version of the data item resides in a cache that is remote relative to the first database server includes the steps of:

determining whether a process currently holds a write lock on the data item; and

if a process currently holds a write lock on the data item, then determining whether the process is remote relative to the first database server.

3. A method for supplying a derived object to a process executing in a node of a plurality of nodes, said plurality of nodes being coupled to a multiple-layer storage system that includes one or more shared storage layers and one or more private storage layers, the method comprising the steps of:

determining derivation data required to build said derived object;

determining where said derivation data is located within said multiple-layer storage system;

selecting a selected node from said plurality of nodes based on where said derivation data is located within said multiple-layer storage system;

causing a derivation mechanism executing on said selected node to build said derived object from said derivation data; and

if said derivation mechanism is not located on the same node as the process, then sending the derived object from said selected node to the node on which the process is executing.

4. The method of claim 3 wherein the step of causing a derivation mechanism executing on said selected node to build said derived object from said derivation data includes the step of sending any reconstruction data that is not currently in the one or more private storage layers of the selected node to said one or more private storage layers of the selected node.

5. The method of claim 3 wherein:

said derived object is derived based on an existing object; and

the step of determining where said derivation data is located within said multiple-layer storage system includes determining where a current version of said existing object is located within said multiple-layer storage system.

6. The method of claim 5 wherein:

the current version of said existing object resides in said one or more private storage layers of a particular node; and

the step of selecting a selected node from said plurality of nodes includes selecting said particular node as said selected node.

7. The method of claim 5 wherein:

the current version of said existing object resides in said one or more shared storage layers; and

the step of selecting a selected node from said plurality of nodes includes selecting the node on which said process resides as said selected node.

8. The method of claim 3 wherein:

portions of said derivation data are stored in said one or more private storage layers on more than one node of said plurality of nodes; and

the method includes the step of determining sizes of said portions of said derivation data; and

the step of selecting a selected node from said plurality of nodes is based on said sizes and where said portions are located in said multiple-layer storage.

9. The method of claim 3 wherein:

a plurality of database servers are executing on said plurality of nodes;

said one or more shared storage layers include one or more shared disks containing a database;

said derived object is a particular version of a data item in said database; and

said derivation data includes a current version of said data item.

10. The method of claim 9 wherein:

the step of determining where said derivation data is located within said multiple-layer storage system includes determining which node of said plurality of nodes currently stores the current version of the data item; and

the step of selecting a selected node includes selecting the node of said plurality of nodes that currently stores the current version of the data item.

11. The method of claim 10 wherein the step of determining which node of said plurality of nodes currently stores the current version of the data item includes determining which database server of said plurality of database servers holds an exclusive mode lock on said data item.

12. A computer readable medium having stored thereon sequences of instructions for supplying a derived object to a process executing in a node of a plurality of nodes, said plurality of nodes being coupled to a multiple-layer storage system that includes one or more shared storage layers and one or more private storage layers, the sequences instructions including sequences of instructions for performing the steps of:

determining derivation data required to build said derived object;

determining where said derivation data is located within said multiple-layer storage system;

selecting a selected node from said plurality of nodes based on where said derivation data is located within said multiple-layer storage system;

causing a derivation mechanism executing on said selected node to build said derived object from said derivation data; and

if said derivation mechanism is not located on the same node as the process, then sending the derived object from said selected node to the node on which the process is executing.

13. The computer readable medium of claim 12 wherein the step of causing a derivation mechanism executing on said selected node to build said derived object from said derivation data includes the step of sending any reconstruction data that is not currently in the one or more private storage layers of the selected node to said one or more private storage layers of the selected node.

14. The computer readable medium of claim 12 wherein:

said derived object is derived based on an existing object; and

the step of determining where said derivation data is located within said multiple-layer storage system includes determining where a current version of said existing object is located within said multiple-layer storage system.

15. The computer readable medium of claim 14 wherein:

the current version of said existing object resides in said one or more private storage layers of a particular node; and

the step of selecting a selected node from said plurality of nodes includes selecting said particular node as said selected node.

16. The computer readable medium of claim 14 wherein:

the current version of said existing object resides in said one or more shared storage layers; and

the step of selecting a selected node from said plurality of nodes includes selecting the node on which said process resides as said selected node.

17. The computer readable medium of claim 12 wherein:

portions of said derivation data are stored in said one or more private storage layers on more than one node of said plurality of nodes; and

the computer readable medium further includes sequences of instructions for performing the step of determining sizes of said portions of said derivation data; and

the step of selecting a selected node from said plurality of nodes is based on said sizes and where said portions are located in said multiple-layer storage.

18. The computer readable medium of claim 12 wherein:

a plurality of database servers are executing on said plurality of nodes;

said one or more shared storage layers include one or more shared disks containing a database;

said derived object is a particular version of a data item in said database; and

said derivation data includes a current version of said data item.

19. The computer readable medium of claim 18 wherein:

the step of determining where said derivation data is located within said multiple-layer storage system includes determining which node of said plurality of nodes currently stores the current version of the data item; and

the step of selecting a selected node includes selecting the node of said plurality of nodes that currently stores the current version of the data item.

20. The computer readable medium of claim 19 wherein the step of determining which node of said plurality of nodes currently stores the current version of the data item includes determining which database server of said plurality of database servers holds an exclusive mode lock on said data item.