Method and mechanism for dependency tracking for unique constraints

Claim 2

A process for tracking dependencies for uniqueness constrained tables in a database system, comprising: identifying a key range in a set of key ranges; maintaining a data structure to hold latest commit time information for said key range; deleting a row from a table at a commit time, said row having a first key value within said key range; updating said data structure based upon said commit time; determining a dependency value for one or more data operations operating upon a second key value in said key range based upon said commit time stored in said data structure; assigning said dependency value to at least one of said one or more data operations operating upon said second key value in said key range; and executing said at least one data operation after deleting said row from said table.

Claim 3

The process of claim 1 in which said second key value is equal to said first key value.

Claim 4

The process of claim 1 in which an index exists with respect to said table and said key range corresponds to a range of index key values.

Claim 5

The process of claim 3 in which said range of index key values is related to an index leaf block.

Claim 6

The process of claim 4 further comprising: splitting said index leaf block into a plurality of index blocks; maintaining a separate data structure for each of said plurality of index blocks, each of said separate data structures having an identical pre-split value from said data structure.

Claim 7

The process of claim 4 further comprising: coalescing a second index leaf block into said index leaf block to form a coalesced index block, said second index block having a second data structure to track commit times deletions from said second index leaf block, said second data structure having a second commit time; and maintaining a single data structure for coalesced index block, said single data structure updated with the later of either the first or second commit times from the first and second data structures.

Claim 8

The process of claim 4 in which said range of index key values comprises a subset of keys in said index leaf block.

Claim 9

The process of claim 4 in which said range of index key values matches a range of keys in said index leaf block.

Claim 10

The process of claim 1 in which said at least one data operation is an insert operation, said insert operation inserting a second row having said second key value in said key range.

Claim 11

The process of claim 1 in which said at least one operation is considered dependent upon one or more data operations that commit on or before said commit time.

Claim 12

The process of claim 1 in which said at least one operation is considered not to be dependent upon any data operation that commits after said commit time.

Claim 13

The process of claim 1 in which said table has a uniqueness constraint upon said key.

Claim 14

A computer program product that includes a computer-usable medium having a sequence of instructions which, when executed by a processor, causes said processor to execute a process for tracking dependencies in a database system, said process comprising: identifying a key range in a set of key ranges; maintaining a data structure to hold latest commit time information for said key range; deleting a row from a table at a commit time, said row having a first key value within said key range; updating said data structure based upon said commit time; determining a dependency value for one or more data operations operating upon a second key value in said key range based upon said commit time stored in said data structure; assigning said dependency value to at least one of said one or more data operations operating upon said second key value in said key range; and executing said at least one data operation after deleting said row from said table.

Claim 15

A computer program product that includes a computer-usable medium having a sequence of instructions which, when executed by a processor, causes said processor to execute a process for tracking dependencies in a database system, said process comprising: identifying a key range in a set of key ranges; maintaining a data structure to hold latest commit time information for said key range; deleting a row from a table at a commit time, said row having a first key value within said key range; updating said data structure based upon said commit time; determining a dependency value for one or more data operations operating upon a second key value in said key range based upon said commit time stored in said data structure; assigning said dependency value to at least one of said one or more data operations operating upon said second key value in said key range; and executing said at least one data operation after deleting said row from said table.

Claim 16

The computer program product of claim 1 in which said second key value is equal to said first key value.

Claim 17

The computer program product of claim 1 in which an index exists with respect to said table and said key range corresponds to a range of index key values.

Claim 18

The computer program product of claim 15 in which said range of index key values is related to an index leaf block.

Claim 19

The computer program product of claim 16 further comprising: splitting said index leaf block into a plurality of index blocks; maintaining a separate data structure for each of said plurality of index blocks, each of said separate data structures having an identical pre-split value from said data structure.

Claim 20

The computer program product of claim 16 further comprising: coalescing a second index leaf block into said index leaf block to form a coalesced index block, said second index block having a second data structure to track commit times deletions from said second index leaf block, said second data structure having a second commit time; maintaining a single data structure for coalesced index block, said single data structure updated with the later of either the first or second commit times from the first and second data structures.

Claim 21

The computer program product of claim 16 in which said range of index key values comprises a subset of keys in said index leaf block.

Claim 22

The computer program product of claim 16 in which said range of index key values matches a range of keys in said index leaf block.

Claim 23

The computer program product of claim 13 in which said at least one data operation is an insert operation, said insert operation inserting a second row having said second key value in said key range.

Claim 24

The computer program product of claim 13 in which said at least one operation is considered dependent upon one or more data operations that commit on or before said commit time.

Claim 25

The computer program product of claim 13 in which said at least one operation is considered not to be dependent upon any data operation that commits after said commit time.

Claim 26

The computer program product of claim 13 in which said table has a uniqueness constraint upon said key.