Restartable database loads using parallel data streams
AssigneeTeradata US, Inc.
InventorsGregory Korenevsky, Alex P. Yung
Life of this patent
Estimating the term…
Abstract
A method and computer program for reducing the restart time for a parallel application are disclosed. The parallel application includes a plurality of parallel operators. The method includes repeating the following: setting a time interval to a next checkpoint; waiting until the time interval expires; sending checkpoint requests to each of the plurality of parallel operators; and receiving and processing messages from one or more of the plurality of parallel operators. The method also includes receiving a checkpoint. request message on a control data stream, waiting to enter a state suitable for checkpointing, and sending a response message on the control data stream.
Claims19
A method for reducing the restart time for a parallel application, the parallel application including a plurality of parallel operators, the method comprising:
- repeating the following by a CRFC component separate from the parallel operators:
- setting a time interval to a next checkpoint;
- waiting until the time interval expires;
- sending checkpoint requests to each of the plurality of parallel operators;
- receiving and processing messages from one or more of the plurality of parallel operators;
- and
- performing the following by the parallel operators:
- receiving a checkpoint request message;
- waiting to enter a state suitable for checkpointing before taking a checkpoint;
- and
- sending a response message.
A method for one of a plurality of parallel operators to record its state, the method comprising the one of the plurality of parallel operators:
- receiving a checkpoint request message on a control data stream;
- waiting to enter a state suitable for checkpointing before taking a checkpoint;
- and
- sending a response message on the control data stream.
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 separate from the parallel operators 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.
Dependencies are inferred from the claim text.