Restartable database loads using parallel data streams

Claim 7

The method of claim 3 further comprising: experiencing a recoverable error; and wherein sending a response message on the control data stream comprises sending a recoverable error message on the control data stream.

Claim 8

The method of claim 3 further comprising: experiencing a non-recoverable error; and wherein sending a response message on the control data stream comprises sending a non-recoverable error message on the control data stream.

Claim 9

A computer program, stored on a tangible storage medium, for use in reducing the restart time for a parallel application, the application comprising a plurality of parallel operators, the computer program comprising: a CRCF component which includes executable instructions that cause a computer to repeat the following; set a time interval to a next checkpoint; wait until the time interval expires; send checkpoint requests to the plurality of parallel operators; receive and process messages from one or more of the plurality of parallel operators; and a plurality of parallel components, each of which is associated with one of the plurality of parallel operators, and each of which includes executable instructions that cause a computer to; receive a checkpoint request message from the CRCF; wait to enter a state suitable for checkpointing; and send a checkpoint response message to the CRCF; wherein each of the parallel components include executable instructions that cause a computer to: determine that the parallel operator is not in a state suitable for checkpointing; and, in sending a response message to the CRCF, the parallel component associated with that parallel operator causes the computer to send a checkpoint reject message to the CRCF; and in receiving and processing messages from one or more of the plurality of parallel operators the CRCF causes the computer to: receive the checkpoint reject message; and send abandon checkpoint messages to the plurality of parallel operators in response to the checkpoint reject message.

Claim 10

A computer program, stored on a tangible storage medium, for use in reducing the restart time for a parallel application, the application comprising a plurality of parallel operators, the computer program comprising: a CRCF component which includes executable instructions that cause a computer to repeat the following; set a time interval to a next checkpoint; wait until the time interval expires; send checkpoint requests to the plurality of parallel operators; receive and process messages from one or more of the plurality of parallel operators; and a plurality of parallel components, each of which is associated with one of the plurality of parallel operators, and each of which includes executable instructions that cause a computer to; receive a checkpoint request message from the CRCF; wait to enter a state suitable for checkpointing; and send a checkpoint response message to the CRCF; wherein each of the parallel components include executable instructions that cause a computer to: determine that the parallel operator is not in a state suitable for checkpointing; and, in sending a response message to the CRCF, the parallel component associated with that parallel operator causes the computer to send a checkpoint reject message to the CRCF; and in receiving and processing messages from one or more of the plurality of parallel operators the CRCF causes the computer to: receive the checkpoint reject message; and send abandon checkpoint messages to the plurality of parallel operators in response to the checkpoint reject message.

Claim 11

A computer program stored on a tangible storage medium, for use in reducing the restart time for a parallel application, the parallel application comprising a plurality of parallel operators, the computer program comprising: a CRCF component which includes executable instructions that cause a computer to repeat the following: set a time interval to a next checkpoint; wait until the time interval expires; send checkpoint requests to the plurality of parallel operators; receive and process messages from one or more of the plurality of parallel operators; and a plurality of parallel components, each of which is associated with one of the plurality of parallel operators, and each of which includes executable instructions that cause a computer to; receive a checkpoint request message from the CRCF; wait to enter a state suitable for checkpointing; and send a checkpoint response message to the CRCF; wherein each of the parallel components include executable instructions that cause a computer to: determine that one or more of the parallel operators has experienced a recoverable error; and, in sending a response message to the CRCF, the parallel component or components associated with the one or more parallel operators that experienced the recoverable error or errors cause the computer to: send a recoverable error message to the CRCF; proceed with recovery; and send a ready message to the CRCF; and in receiving and processing messages from one or more of the plurality of parallel operators, the CRCF causes the computer to: receive the recoverable error message; send abandon checkpoint messages to the plurality of parallel operators in response to the recoverable error message; wait for the ready messages; receive the ready messages; and schedule a checkpoint.

Claim 12

A computer program stored on a tangible storage medium, for use in reducing the restart time for a parallel application, the parallel application comprising a plurality of parallel operators, the computer program comprising: a CRCF component which includes executable instructions that cause a computer to repeat the following: set a time interval to a next checkpoint; wait until the time interval expires; send checkpoint requests to the plurality of parallel operators; receive and process messages from one or more of the plurality of parallel operators; and a plurality of parallel components, each of which is associated with one of the plurality of parallel operators, and each of which includes executable instructions that cause a computer to; receive a checkpoint request message from the CRCF; wait to enter a state suitable for checkpointing; and send a checkpoint response message to the CRCF; wherein each of the parallel components include executable instructions that cause a computer to: determine that one or more of the parallel operators has experienced a recoverable error; and, in sending a response message to the CRCF, the parallel component or components associated with the one or more parallel operators that experienced the recoverable error or errors cause the computer to: send a recoverable error message to the CRCF; proceed with recovery; and send a ready message to the CRCF; and in receiving and processing messages from one or more of the plurality of parallel operators, the CRCF causes the computer to: receive the recoverable error message; send abandon checkpoint messages to the plurality of parallel operators in response to the recoverable error message; wait for the ready messages; receive the ready messages; and schedule a checkpoint.