Mechanism for managing execution of interdependent aggregated processes

A computer implemented method for managing execution of work in a networked multi-processor computing environment in which work is distributed between multiple processors, the method comprising:

• receiving a work request, said work request specifying a first job and a plurality of other jobs, said first job comprising a first set of one or more sub-works, said first job having a dependency which is resolved via a logical expression, said logical expression having as operands the states of said plurality of other jobs;
• processing said work request to automatically determine that said first job has said dependency;
• placing each of said plurality of other jobs into an active state;
• placing said first job into a pending state;
• determining the states of said plurality of other jobs;
• resolving said logical expression using said states to determine whether said dependency has been met;
• in response to a determination that said dependency has been met, placing said first job into an active state;
• and
• scheduling said first set of sub-works for execution;
• wherein said first set of sub-works is not scheduled for execution until after said first job is placed in an active state.

A computer-readable medium carrying one or more sequences of instructions for managing execution of work in a networked multi-processor computing environment in which work is distributed between multiple processors, wherein execution of the one or more sequences of instructions by one or more processors causes the one or more processors to perform the steps of:

• receiving a work request, said work request specifying a first job and a second job, said first job comprising a first set of one or more sub-works and said second job comprising a second set of one or more sub-works, said second job having a dependency on said first job meeting a particular condition;
• wherein said first and second jobs comprise respective aggregations of executable commands and wherein said first and second sub-works comprise at least a portion of said executable commands;
• interpreting said work request to automatically determine that said second job has said dependency on said first job and to automatically determine that said second sub-works has said dependency on said first job;
• placing said first job into an active state to enable said first set of sub-works to be scheduled for execution;
• placing said second job into a pending state;
• determining whether said first job has met said particular condition;
• in response to a determination that said first job has met said particular condition, thereby indicating that said dependency has been met, automatically placing said second job into an active state;
• and
• in response to said second job being placed into an active state, scheduling for execution on said networked multi-processor computing environment said second set of sub-works, without yet sending said second set of sub-works to said networked multi-processor computing environment for execution.

A computer-readable medium carrying one or more sequences of instructions for managing execution of work in a networked multi-processor computing environment in which work is distributed between multiple processors, wherein execution of the one or more sequences of instructions by one or more processors causes the one or more processors to perform the steps of:

• receiving a work request, said work request specifying a first job and a plurality of other jobs, said first job comprising a first set of one or more sub-works, said first job having a dependency which is resolved via a logical expression, said logical expression having as operands the states of said plurality of other jobs;
• processing said work request to automatically determine that said first job has said dependency;
• placing each of said plurality of other jobs into an active state;
• placing said first job into a pending state;
• determining the states of said plurality of other jobs;
• resolving said logical expression using said states to determine whether said dependency has been met;
• in response to a determination that said dependency has been met, placing said first job into an active state;
• and
• scheduling said first set of sub-works for execution;
• wherein said first set of sub-works is not scheduled for execution until after said first job is placed in an active state.

A network device that can manage execution of work in a networked multi-processor computing environment in which work is distributed between multiple processors, the network device comprising:

• a network interface;
• a processor coupled to the network interface and receiving network messages from the network through the network interface;
• a computer-readable medium comprising one or more stored sequences of instructions which, when executed by the processor, cause the processor to carry out the steps of:
• receiving a work request, said work request specifying a first job and a second job, said first job comprising a first set of one or more sub-works and said second job comprising a second set of one or more sub-works, said second job having a dependency on said first job meeting a particular condition;
• wherein said first and second jobs comprise respective aggregations of executable commands and wherein said first and second sub-works comprise at least a portion of said executable commands;
• interpreting said work request to automatically determine that said second job has said dependency on said first job and to automatically determine that said second sub-works has said dependency on said first job;
• placing said first job into an active state to enable said first set of sub-works to be scheduled for execution;
• placing said second job into a pending state;
• determining whether said first job has met said particular condition;
• in response to a determination that said first job has met said particular condition, thereby indicating that said dependency has been met, automatically placing said second job into an active state;
• and
• in response to said second job being placed into an active state, scheduling for execution on said networked multi-processor computing environment said second set of sub-works, without yet sending said second set of sub-works to said networked multi-processor computing environment for execution;
• wherein said sub-job is not scheduled for execution until said second job is placed in an active state.

A network device that can manage execution of work in a networked multi-processor computing environment in which work is distributed between multiple processors, the network device comprising:

• a network interface;
• a processor coupled to the network interface and receiving network messages from the network through the network interface;
• a computer-readable medium comprising one or more stored sequences of instructions which, when executed by the processor, cause the processor to carry out the steps of:
• receiving a work request, said work request specifying a first job and a plurality of other jobs, said first job comprising a first set of one or more sub-works, said first job having a dependency which is resolved via a logical expression, said logical expression having as operands the states of said plurality of other jobs;
• processing said work request to automatically determine that said first job has said dependency;
• placing each of said plurality of other jobs into an active state;
• placing said first job into a pending state;
• determining the states of said plurality of other jobs;
• resolving said logical expression using said states to determine whether said dependency has been met;
• in response to a determination that said dependency has been met, placing said first job into an active state;
• and
• scheduling said first set of sub-works for execution;
• wherein said first set of sub-works is not scheduled for execution until after said first job is placed in an active state.