Distributed workflow device and method for processing workflow by distributed workflow device

Abstract

The embodiment of the invention provides a distributed workflow device and a method for processing workflow by the distributed workflow device, wherein the distributed workflow device concretely comprises a main node and state nodes in a cluster; the main node is used for allocating one corresponding state node to a workflow definition submitted by a user; the workflow definition provides a task interface, and the task interface can be used for calling a sub system for executing a corresponding workflow task; each state node is used as a management calling node for compiling the workflow definition into the workflow task and the corresponding task interface, and distributing the workflow task and the corresponding task interface to the corresponding state node; and the state nodes are also used as task executing nodes for executing the workflow task by the corresponding sub system called by the task interface. The distributed workflow device and the method provided by the embodiment have the universality, and the workflow processing stability and throughput can be improved.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a distributed workflow device and a method for processing workflow. Background technique [0002] Workflow is the abstraction, generalization and description of business rules between workflow and its various operation steps. Workflow modeling refers to expressing the logic and rules of how the work in the workflow is organized together with an appropriate model in the computer and performing calculations on it. Workflow modeling is usually done by a workflow engine. [0003] In the large-scale Internet content production system, there are many production systems similar to workflows. For example, a video publishing process generally includes: transcoding, non-editing, uploading, publishing and other processes. The start of a process depends on the end status of the previous process, thus forming a typical video production workflow. [0004] Unlike traditional workflows, eac...

AI Technical Summary

Problems solved by technology

[0006] First, the versatility is poor; the general customization system is customized for a certain characteristic requirement, and the customization work mainly includes the interface between various subsystems, the communication protocol of the message system, etc. If there is a new requirement, the new requirement cannot be implemented. Reuse existing interfaces and protocols

[0007] Second, the stability is poor; because the communication information between systems is transmitted through the message system, the scalability of the high-reliability message system is usually not good, and once the message is backlogged, the entire message system is likely to be unavailable, which in turn leads to the entire workflow. system block

[0008] Third, there is a lack of unified control; each subsystem operates independently, and the only connection is to communicate through the message system, so it is impossible to understand the overall running status of the workflow

[0009] Fourth, it is not easy to maintain; the maintenance and upgrade of the system needs to be coordinated by each subsystem, and there is no unified operation interface, which will generate additional operation and maintenance burden

Images