转自:https://blog.bernd-ruecker.com/bizdevops-the-true-value-proposition-of-workflow-engines-f342509ba8bb

Whenever you implement core business logic you have to discuss it with a lot of stakeholders and having a graphical representation is a huge help in doing so. This is why there are a lot of methodologies out there to capture business logic visually, like Event Stormingdomain modeling and of course process modeling with BPMN.

BPMN is an ISO standard and its beauty is that it defines a business readable graphical visualization that is also directly executable on a workflow engine. A BPMN model therefore is also an XML file a workflow engine can understand.

Compared to other languages, BPMN is a good choice as it is really widespread and very mature. I will not dive into details of the BPMN elementsbut I can report that it was sufficient for all real-life use cases we saw in the last years — which are a lot.

It is about living documentation — not low code!

The benefit of directly executable models is what Cyrille Martraire named living documentation. The model is not an input of the requirement phase that will be outdated once the first line of code is written. No, the model will be executed and is the code. It is always in-sync with the implementationduring the whole life-cycle.

Unfortunately this is often connected with zero or low code approaches which try to get rid of the developer. In our experience these approaches do not work well, at least not in non-trivial use cases. I see BPMN models simply as a different way of writing code which should still be done by a developer. Of course he/she communicates with business stakeholders to do the right thing. And the true benefit of using BPMN and a workflow engine is to facilitate the collaboration between the different roles!

BizDevOps

Having an executable, visual and understandable model has benefits for business stakeholders, developers and operators. It also improves the communication and collaboration between them big time.

Let’s dive into that:

The business can understand visual workflows and discuss the underlying business process and related requirements.

Graphical report for one scenario run as automated unit test (here: Camunda Test Coverage). The payment workflow is taken from the flowing retail example. Use cases for workflows can be found in any industry.

Developers get a workflow engine which is a persistent state machine with a lot of additional features. On top of that they can benefit from the visual model as they gain more context into what they are actually working on. For example, some tools can create graphical reports for test scenarios.

Exemplary operations tool from a workflow engine (here: Camunda Cockpit)

Another huge gain lies within operations as a workflow engine provides a lot of context to understand and resolve problems or failures. The possibilities to directly intervene in real-time (e.g. repair failed workflow instances or correct data) help keep everything up and running.

Audit data can be leveraged to optimize workflows (here: Camunda Optimize)

And as workflow engines write a lot of audit data, one can leverage this historical data to find bottlenecks and possibilities for improvement for your workflows. This might be done by any stakeholder as everybody can understand the BPMN workflow models.

Summary

In this short post I wanted to highlight the value of graphical but directly executable models in BPMN for business stakeholders, developers and operators (BizDevOps). This is not about getting rid of IT by following a low code approach but rather provide benefits to each of the roles and facilitate collaboration between them.

PS: Screenshots are taken from Camunda, an open source platform for workflow automation. As co-founder of Camunda I simply know this tool the best.

 
 
 
 

BizDevOps — the true value proposition of workflow engines的更多相关文章

  1. The Microservices Workflow Automation Cheat Sheet

    Written by Bernd Rücker on Dec 12 2018 in the Best Practices category. Editor’s Note: This post orig ...

  2. How to implement long running flows, sagas, business processes or similar

    转自:https://blog.bernd-ruecker.com/how-to-implement-long-running-flows-sagas-business-processes-or-si ...

  3. Architecture options to run a workflow engine

    This week a customer called and asked (translated into my own words and shortened): “We do composite ...

  4. Rule Or WorkFlow

    The main value of a Workflow engine is that it makes it possible to customize the flows through some ...

  5. 风尘浪子 只要肯努力,梦想总有一天会实现 WF工作流与Web服务的相互调用 —— 通过Web服务调用Workflow工作流(开发持久化工作流) _转

    如果你曾经负责开发企业ERP系统或者OA系统,工作流对你来说一定并不陌生.工作流(Workflow)是对工作流程及其各操作步骤之间业务规则 的抽象.概括.描述.工作流要解决的主要问题是:为实现某个业务 ...

  6. 中文分词工具探析(二):Jieba

    1. 前言 Jieba是由fxsjy大神开源的一款中文分词工具,一款属于工业界的分词工具--模型易用简单.代码清晰可读,推荐有志学习NLP或Python的读一下源码.与采用分词模型Bigram + H ...

  7. Oozie_示例

    Oozie 官方示例 解压oozie-examples.tar.gz 将examples/上传到HDFS家目录 $ bin/hdfs dfs -put /opt/cdh-5.6.3/oozie-4.0 ...

  8. Odoo ir actions 分析

    源代码位置:openerp/addons/base/ir/ir_actions.py 根类型:ir.actions.actions class actions(osv.osv): _name = 'i ...

  9. WWF3XOML方式创建和启动工作流 <第十篇>

    一.XOML使用工作流的好处 通过Xoml方式使用工作流的好处在于,它能够不重新启动程序的情况下,仅仅通过配置xoml就能够实现改变工作流,非常灵活. 创建一个WinForm程序如下: 代码如下: n ...

随机推荐

  1. 8.1 C++输入输出类的层次

    参考:http://www.weixueyuan.net/view/6407.html 总结: 在C++中,输入输出数据的传送过程我们称之为流,一个流就是一个字节序列,对流可以进行读或写操作. 输入输 ...

  2. 网口扫盲二:Mac与Phy组成原理的简单分析(转)

    1. general 下图是网口结构简图.网口由CPU.MAC和PHY三部分组成.DMA控制器通常属于CPU的一部分,用虚线放在这里是为了表示DMA控制器可能会参与到网口数据传输中. 对于上述的三部分 ...

  3. U盘如何去除写保护

    用一个简单而粗暴的方法去除U盘写保护: www.upantool.com/hfxf/xiufu//FormatTool.html#softdown 没有比这个更牛的软件了.

  4. AngularJS2.0教程(一)快速上手之基础知识

    Why Angular2 Angular1.x显然非常成功,那么,为什么要剧烈地转向Angular2? 性能的限制 AngularJS当初是提供给设计人员用来快速构建HTML表单的一个内部工具.随着时 ...

  5. POJ 2407 Relatives(欧拉函数入门题)

    Relatives Given n, a positive integer, how many positive integers less than n are relatively prime t ...

  6. <算法><Union Find并查集>

    Intro 想象这样的应用场景:给定一些点,随着程序输入,不断地添加点之间的连通关系(边),整个图的连通关系也在变化.这时候我们如何维护整个图的连通性(即判断任意两个点之间的连通性)呢? 一个比较简单 ...

  7. jetty调优

    jetty服务器使用遇到一下内存溢出的问题: java.lang.OutOfMemoryError: unable to create new native thread   无法创建新的进程 方法: ...

  8. 利用python将数据转存入sqlite3

    案例的目标是将存在文件中的json格式数据转存到sqlite数据库中.因此,需要利用python逐行读取json文件中数据,对数据进行解析和入库.具体操作步骤如下: 1.逐行读取json文件 for ...

  9. CRM工具简介

    pacemaker是高可用集群中的CRM(Cluster Resource Manager)资源管理层,他是一个服务,可以作为一个单独的服务启动,不过在如果使用corosync1.4中,我们可以设置c ...

  10. HDU 6098 17多校6 Inversion(思维+优化)

    Problem Description Give an array A, the index starts from 1.Now we want to know Bi=maxi∤jAj , i≥2. ...