Webinar: 7 Habits of the Successful Boomi Developer

Video created by Adam Arrowsmith Employee on Feb 28, 2018

    Boomi’s visual, drag and drop UI, along with our unified platform, makes it easy to build, deploy and manage integration processes. The question is – are you effectively using the power of the Boomi platform to drive productivity across your development teams?


    Hear from Adam Arrowsmith, Solutions and Community Manager at Dell Boomi and Brian Hausmann, Integration Specialist at Cloudera, as they provide practical development strategies and configuration tips for both the novice and the experienced Boomi developer to start using today!


    In this webinar you will learn:

    • How to establish development best practices across your team(s)
    • Strategies for approaching new integrations
    • Configuration tips for building manageable integrations


    Here are the questions and answers asked at the end of the session. If there are any additional questions or clarifications, please post a comment below.



    Q: Is the document cache available on the basic version of AtomSphere?No. The Document Cache component is part of Advanced Workflow, available in the Professional Edition or higher currently.
    Q: Hi, What is the best way to get document tracking in low latency mode?You can't. One of the trade offs for achieving the high volume throughput of real time requests in Low Latency mode is to disable tracking of individual executions, which includes document-level tracked fields.
    Q: Do you have to deploy Process Route before you can test it? For example, if we just want to test out an idea without deploying anything first, is this possible using Process Route too. Nope, the Process Route shape is Test Mode-able. See Process testing.
    Q: Boomi Flow rocks!Thanks, we think so too! Check out https://docs.manywho.com/ to get started.
    Q: Can we automate the deloyments using boomi Atomsphere for continous integration and continous delivery or how do we automate the deployments?Yes. The AtomSphere API allows you to deploy and copy deployments across environments. Check out Automating Deployments with the AtomSphere API.
    Q: does boomi have a professional services arm if you don't have expertise in house?Absolutely! Our professional services team offers the complete range of services, from enterprise architecture/strategy to project delivery to staff augmentation. Additionally, our Boomi U training is freely available online or instructor led (paid). We also have a network of implementation partners to assist as well.
    Q: How long would it take to train a staffer? What would be their core competency have to be (programmer)?

    We offer self-paced, online training (free) as well as paid, instructor-led courses that typically span 2-3 days. Many individuals are up and running with their first simple integration in a very short time.


    Because Boomi is a graphical, configuration-based tool, programming skills are not required. A solid understanding of your applications, their data and business process are more important than coding skills. However individuals with a development/technical background typically pick it up even faster.

    Q: Did Cloudera find any useful connections or mappings in Ideas or the community?While Brian is active in the community and participates in discussions and ideas, he did not leverage any specific mappings from the community or Process Library examples for their projects to date.
    Q: In process routing, can you rerun some documents at the runtime thru the subprocess without initiating them from the parent process?No. From an execution perspective, the Process Route subprocesses behave the same as normal Process Calls. To achieve that, you would need to adopt a decoupled design via web services or message queuing.
    Q: Is the transaction management can be done with Boomi code , I mean with Commit and Rollback ? Any best practices with transaction management ?

    Transaction management (e.g. XA) is very often associated with database and messaging architectures and is not really present in distributed, web services architectures. That said, the Database connector provides commit/rollback options when you have multiple SQL Statements in a single profile. It does not provide transaction control across multiple connector calls in a process.


    Check out this idea and this discussion and this discussion.

    Q: Document Tracking tips/use case for troubleshooting?Check out Document Tracking Best Practices.
    Q: How do you document your integration for your Functional Business System Analyst to understand?

    We find a written document (e.g. Word doc) inclusive of the following aspects most helpful: short overview, key assumptions, filter criteria, business logic/routing/validations, and field mapping (exported from the map component).


    Additionally a screen shot of the process flow with numbering and footnotes for key shapes or sections of the process are helpful.

    Q: Is there any more advanced training regarding Process Routes?Check out How to use the Process Route shape and component.
    Q: The dynamic process property is persisted in the memory? when the Local Atom service is restarted will it reset the dpp value as well?

    By default process property or dynamic process property values live in memory for the life of a given execution. However it is possible to explicitly persist a value to remember it across executions. This would include Atom restarts.


    For a practical example of this, check out How to sync records incrementally by Last Modified Date.

    Q: Is there an option to runs sub process in parallel from Main process instead of Serial?

    Not currently with simple shape configuration but it can be achieved through process design. Check out Design Pattern: How to Create a Web Service Mashup/Composite Call with an Aggregated Response.


    See also Ability to process the same documents down multiple paths simultaneously.

    Q: how do we achieve the version controlTo better understand versioning within AtomSphere, check out Boomi U Architecture courses. A slightly older version of much of the same content/concepts is available here: Managing the AtomSphere Development Life Cycle Part 1 - Development.
    Q: Docker to install an Atom - how can we use it for scaling?NEED TO GET MORE INFO...WILL UPDATE SHORTLY. FOLLOW THIS DOCUMENT TO BE NOTIFIED OF UPDATES.
    Q: You talked about using cache when looking up data via the API. Can it be used with sql lookups on local atoms?No, the SQL Lookup map function or Program Command shape option do not support caching in that way. That would be a good opportunity to use the Document Cache component approach (check out Document Cache Best Practices and Common Scenarios).
    Q: How do you increase process execution history retention period (I believe the default is 30 days)?Currently execution history metadata (what you see in Process Reporting) cannot be extended beyond 30 days. This is a common feedback item and the product team is aware of it, but there is no definitive roadmap item for extending the retention period at this time. Vote/comment on idea  Process Reporting for more than 30 days.
    Q: Best practices/ considerations while migrating from other Integration tool to Dell Boomi.Check out Approach for Migrating from an Existing Middleware Platform to AtomSphere.
    Q: What is the best way to implement continuous integration and test automation in Boomi?

    While the Deployment API discussed above can help automate deployments into a dev/staging environment, there are some considerations and best practices discussed in that article linked above.


    As a starting point, check out Testing Strategies. Test automation (e.g., unit, regression) is an area we are looking to get additional feedback on--please share ideas/use cases here:  Unit Test Component? 

    Q: What to consider for disaster recovery?Check out Best Practices for Run Time High Availability and Disaster Recovery.
    Q: Is there any plan to modernize the branching and code management in Boomi?

    As a high-productivity, configuration-based tool, many of the traditional version control approaches (and consequently the complexity they bring) for text-based development are not always applicable. That said, we do recognize some of the practical management challenges that arise when the number concurrent developers and shared components increase. We are working on some significant deployment versioning changes which lay some foundational elements for additional capabilities to potentially address some of the advanced version management scenarios.


    Also check out this idea/discussion: Code merge / Advance Copy Operation.



    Mariam Daud