Serverless computing and task scheduling

Patent metadata

Field	Value
Publication number	US-10884807-B2
Application number	US-201715485910-A
Country	US
Kind code	B2
Filing date	Apr 12, 2017
Priority date	Apr 12, 2017
Publication date	Jan 5, 2021
Grant date	Jan 5, 2021

Abstract

In one embodiment, a method for serverless computing comprises: receiving a task definition, wherein the task definition comprises a first task and a second task chained to the first task; adding the first task and the second task to a task queue; executing the first task from the task queue using hardware computing resources in a first serverless environment associated with a first serverless environment provider; and executing the second task from the task queue using hardware computing resources in a second serverless environment selected based on a condition on an output of the first task.

First claim

What is claimed is:

1. A method for serverless computing, comprising:
receiving a task definition, wherein the task definition comprises a first task and a second task chained to the first task;
adding the first task and the second task to a task queue;
determining a set of dependency constraints for each task of a task chain in the task queue;
determining a cost and a set of capabilities associated with each serverless computing environment;
executing the first task from the task queue using hardware computing resources in a first serverless environment associated with a first serverless environment provider;
first selecting, based on sets of dependency constraints, the costs, the sets of capabilities, and an output of the first task, a second serverless environment out of a plurality of serverless environments to execute the second task from the task queue; and
executing the second task from the task queue using hardware computing resources in the second serverless environment.

2. The method of claim 1 , wherein the task definition comprises a first task identifier identifying the first task, a first pointer to input data, a task action function code, a second pointer to output data, and a second task identifier identifying the second task.

3. The method of claim 1 , wherein the first selecting the serverless computing environment comprises minimizing sum of costs for all tasks in the task chain and ensuring the set of dependency constraints for each task in the task chain is satisfied by the set of capabilities associated with the serverless environment selected for each task in the task chain.

4. The method of claim 1 , wherein the set of dependency constraints comprises code for the task, and the set of capabilities comprises a programming language that a serverless environment can execute.

5. The method of claim 1 , wherein a set of dependency constraints for a given task comprises a data locality compliance rule.

6. The method of claim 1 , wherein a set of dependency constraints for a given task comprises one or more requirements specified in a task definition associated with the given task.

7. The method of claim 1 , wherein a set of dependency constraints for a given task comprises a rule specifying a particular serverless environment to be used for the given task if a condition of an output of a task previous to the given task in the task chain is met.

8. A system comprising:
at least one memory element;
at least one processor coupled to the at least one memory element;
an interface that when executed by the at least one processor is configured to: receive a task definition, wherein the task definition comprises a first task and a second task chained to the first task and adding the first task and the second task to a task queue; and
one or more workers provisioned in networked hardware resources of a serverless computing environment that when executed by the at least one processor is configured to:
determine a set of dependency constraints for each task of a task chain in the task queue;
determine a cost and a set of capabilities associated with each serverless computing environment;
execute the first task from the task queue using hardware computing resources in a first serverless environment associated with a first serverless environment provider;
first select, based on a sets of dependency constraints, the costs, and the sets of capabilities, and an output of the first task, a second serverless environment out of a plurality of serverless environments to execute the second task from the task queue; and
execute the second task from the task queue using hardware computing resources in the second serverless environment.

9. The system of claim 8 , wherein the task definition comprises a first task identifier identifying the first task, a first pointer to input data, a task action function code, a second pointer to output data, and a second task identifier identifying the second task.

10. One or more computer-readable non-transitory media comprising one or more instructions, for serverless computing and task scheduling, that when executed on a processor configure the processor to perform one or more operations comprising:
receiving a task definition, wherein the task definition comprises a first task and a second task chained to the first task;
adding the first task and the second task to a task queue;
determining a set of dependency constraints for each task of a task chain in the task queue;
determining a cost and a set of capabilities associated with each serverless computing environment;
executing the first task from the task queue using hardware computing resources in a first serverless environment associated with a first serverless environment provider;
first selecting, based on sets of dependency constraints, the costs, and the sets of capabilities, and an output of the first task, a second serverless environment out of a plurality of serverless environments to execute the second task from the task queue; and
executing the second task from the task queue using hardware computing resources in the second serverless environment.

11. The media of claim 10 , wherein the first selecting the serverless computing environment comprises minimizing sum of costs for all tasks in the task chain and ensuring the set of dependency constraints for each task in the task chain is satisfied by the set of capabilities associated with the serverless environment selected for each task in the task chain.

12. The media of claim 10 , wherein the set of dependency constraints comprises the programming language of code for the task, and the set of capabilities comprises a programming language that a serverless environment can execute.

13. The media of claim 10 , wherein a set of dependency constraints for a given task comprises one or more of the following: a data locality compliance rule, and one or more requirements specified in a task definition associated with the given task.

14. The media of claim 10 , wherein a set of dependency constraints for a given task comprises a rule specifying a particular serverless environment to be used for the given task if a condition of an output of a task previous to the given task in the task chain is met.

Assignees

• Cisco Tech Inc

Inventors

• Shimamura Komei
• Okwii Timothy
• Dutta Debojyoti
• Udupi Yathiraj B
• Ramakrishna Rahul
• Huang Xinyuan

Classifications

• G06F9/5027 Primary

  the resource being a machine, e.g. CPUs, Servers, Terminals · CPC title

• G06F9/5066

  Algorithms for mapping a plurality of inter-dependent sub-tasks onto a plurality of physical CPUs (mappping at compile time, see G06F8/451) · CPC title