Automated message-based job flow resource management in container-supported many task computing

Patent metadata

Field	Value
Publication number	US-11086608-B2
Application number	US-202017139503-A
Country	US
Kind code	B2
Filing date	Dec 31, 2020
Priority date	Feb 5, 2016
Publication date	Aug 10, 2021
Grant date	Aug 10, 2021

Abstract

An apparatus includes at least one processor to: within a kill container, execute a kill routine to monitor a task kill queue for storage of an execution status message indicating a level of a parameter of execution of a task routine to perform a task of a job flow, and in response to the level exceeding a threshold, store, within the task kill queue, a kill tasks request message; within a task container, in response to the kill tasks request message, cease execution of the task routine, and store, within a task queue, a task cancelation message; within a performance container, execute instructions of a performance routine to, in response to the task cancelation message, store, within a job queue, a job cancelation message; and in response to the job cancelation message, transmit an indication of cancelation of the job flow to a requesting device.

First claim

The invention claimed is:

1. An apparatus comprising at least one processor and a storage to store instructions that, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
within a first kill container, execute instructions of a first instance of a kill routine to cause the at least one processor to perform operations comprising:
monitor a task kill queue for storage of an execution status message within the task kill queue that indicates a level of a parameter of execution of a first task routine of a set of task routines to perform a first task of a set of tasks of a job flow, wherein:
the job flow is defined in a job flow definition that specifies the set of tasks to be performed via execution of the set of task routines during performance of the job flow;
the job flow definition is stored among multiple job flow definitions within at least one federated area;
the set of task routines is stored among multiple task routines within the at least one federated area; and
the at least one federated area is maintained within at least one storage device;

compare the level of the parameter to a threshold limit level for the parameter; and
in response to the level of the parameter exceeding the threshold limit level, store, within the task kill queue, a kill tasks request message comprising a job flow identifier of the job flow;
within a first task container, in response to the storage of the kill tasks request message within the task kill queue, perform operations comprising:
cease execution of instructions of the first task routine to cease performance of the first task; and
store, within a task queue, a first task cancelation message indicative of cancelation of execution of the first task routine;

within a second task container, in response to the storage of the kill tasks request message within the task kill queue, perform operations comprising:
cease execution of instructions of a second task routine of the set of task routines to cease performance of a second task of the set of tasks; and
store, within the task queue, a second task cancelation message indicative of cancelation of execution of the second task routine;

within a first performance container, execute instructions of a first instance of a performance routine to cause the at least one processor to, in response to the storage of either the first task cancelation message within the task queue or the second task cancelation message within the task queue, perform operations comprising:
store, within a job queue, a job cancelation message indicative of cancelation of the job flow; and
in response to the storage of the job cancelation message within the job queue, transmit an indication of cancelation of the job flow to a requesting device via a network.

2. The apparatus of claim 1 , wherein the at least one processor is caused to perform operations comprising:
within a second kill container, execute instructions of a second instance of the kill routine to cause the at least one processor to monitor the task kill queue for storage, within the task kill queue, of the execution status message indicating the level of the parameter of execution of the first task routine;
prior to the storage of the execution status message within the task kill queue, make a first determination, based on availability of at least one of processing resources and storage resources, that the second kill container is to be uninstantiated;
in response to the first determination, uninstantiate the second kill container;
at a time following the uninstantiation of the second kill container, make a second determination, based on the availability of at least one of the processing resources and the storage resources, that the first kill container is to be instantiated; and
in response to the second determination, instantiate the first kill container.

3. The apparatus of claim 1 , wherein:
within the first performance container, in response to storage, in the job queue, of a job performance request message that conveys a request to perform the job flow, a first task routine execution request message and a second task routine execution request message were stored within the task queue and the task queue was monitored for indications of completion of execution of the set of task routines;
within the first task container, execution of the instructions of the first task routine commenced within the first task container in response to the storage of the first task routine execution request message within a task queue;
within the second task container, execution of the instructions of the second task routine commenced within the second task container in response to storage of the second task routine execution request message within the task queue; and
the at least one processor is caused to perform operations comprising:
within the first performance container, the at least one processor is caused to, in response to the storage of either the first task cancelation message within the task queue or the second task cancelation message within the task queue, perform operations comprising:
cease monitoring of the task queue for indications of completion of execution of the set of task routines.

4. The apparatus of claim 1 , wherein:
the job flow definition specifies data dependencies among the set of tasks; and
the at least one processor is caused to perform operations comprising:
receive, via the network, a request to perform the job flow from the requesting device;
retrieve the job flow definition from among the multiple job flow definitions in the at least one federated area;
store, within the job queue, a job performance request message comprising the job flow definition;
within the first performance container, in response to the storage of the job performance request message within the job queue, perform operations comprising:
based on the data dependencies specified in the job flow definition, derive an order of performance of the set of tasks that specifies at least the first task of the set of tasks to be performed;
store, within the task queue, a first task routine execution request message comprising an identifier associated with the first task; and
monitor the task queue for an indication of completion of execution of the first task routine; and

within the first task container, in response to the storage of the first task routine execution request message within the task queue, perform operations comprising:
use the identifier associated with the first task to retrieve the first task routine from the at least one federated area; and
commence execution, within the first task container, of instructions of the first task routine to commence performance of the first task.

5. The apparatus of claim 4 , wherein:
the request to perform a job flow comprises the job flow identifier of the job flow definition;
the retrieval of the job flow definition comprises use of the job flow identifier, wherein the job flow definition comprises a set of flow task identifiers that each identify a task of the set of tasks;
the identifier associated with the first task comprises a flow task identifier of the set of flow task identifiers that identifies the first task; and
the retrieval of the first task routine comprises selecting, from among the multiple task routines stored within the at least one federated area, a latest version of task routine comprising instructions that, when executed by the at least one processor, causes the at least one processor to perform the first task.

6. The apparatus of claim 4 , wherein:
the job flow definition specifies data dependencies among the set of tasks; and
the at least one processor is caused to perform operations comprising:
within

Assignees

• Sas Inst Inc

Inventors

• Bequet Henry Gabriel Victor
• Stogner Ronald Earl
• Yang Eric Jian
• Gong Qing

Classifications

• G06N3/043

  based on fuzzy logic, fuzzy membership or fuzzy inference, e.g. adaptive neuro-fuzzy inference systems [ANFIS] · CPC title

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title

• G06N3/048

  Activation functions · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06F9/5038

  considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration (scheduling strategies G06F9/4881 and subgroups) · CPC title