Just recently I wrote about why Extreme is Going Serverless and the multitude of benefits for developers in an environment with multiple clouds and on-premises solutions, as well as numerous apps. Serverless saves cost and resources, particularly in a multi-cloud environment.
Last week at ServerlessConf in San Francisco we introduced our new workflow engine Orquesta (Spanish for orchestra) with demos showcasing it as a new multi-cloud serverless orchestrator. You can readily start using it in your solutions – it’s open source and downloadable from Github, or you can sign up to a preview of a hosted version – at orquesta.cloud.
Imagine you are running a large healthcare organization and you have a set of data that needs to be analyzed. This statistical analysis is however computing intensive and you just don’t have that computing power on site. Or, you don’t have time to set up a Hadoop cluster, optimize it, and then install Apache Spark, Logstash, and TensorFlow along with all the tooling to keep it running and analyze your large set of data. Or you want to take advantage of machine learning and AI offerings from Azure, AWS, or Google Cloud Composer and for that you need to upload your data.
No matter what your reasons are, you end up uploading your dataset to the cloud of choice – or even multiple clouds – and use the best of breed PaaS services to get it analyzed. Once that’s done, you receive the results and remove the data from the cloud to stay compliant.
Since your data analysts need this solution for analyzing large data sets regularly, you want to automate the entire workflow. That’s where our Orquesta comes in.
Our new workflow engine boasts a DSL – workflow definition language – that supports more patterns than Azure Logic Apps or AWS Step Functions, yet is simple and expressive. With this DSL you can define orchestrations of tasks across multiple clouds and on-premises systems. You can embed Orquesta in your own solution as a library, run it as part of StackStorm or upgrade to our commercial Workflow Composer for visual workflow composition and enterprise support. Your orchestrations can call serverless functions or API endpoints in major clouds, or use actions from StackStorm Exchange for more elaborate integrations. The possibilities are open to your skills and imagination.
Dmitri Zimine, co-founder of StackStorm, describes the difference between the three major clouds that have introduced Workflow-as-a-Service offerings: Microsoft Azure, Amazon Web Services, and the Google Cloud Platform in his article Serverless and Workflows: The Present and the Future. These major cloud providers each have their own workflow services with different concepts, languages, and integration packs, and with that they lock you into a specific cloud. They are equally capable, and with enough time and ingenuity any developer can achieve the same results. However, only Google Cloud Composer, can be used on-prem as it’s based on open source Apache AirFlow. Comparing all three WFaaS offerings, none of them shows an advantage for multi-cloud orchestration.
You can write workflows and logic with StackStorm and apply them across cloud providers. When you take advantage of different cloud platforms, often you run into the problem that a certain service can only run on one specific cloud. Another service is only available on another cloud and it’s hard to make the different cloud platforms talk to each other. This is where StackStorm with Orquesta can help you automate and integrate across your mixed infrastructure, on-premises, private/hybrid/public cloud. Orquesta allows you to take advantage of the best of all of them and let them communicate with each other.
If you decide to take a serverless route, don’t host or operate a workflow orchestrator, and only pay per use – sign up for orquesta.cloud. This will bring the power of Orquesta, the convenience of visual workflow composition, and the flexibility of multi-cloud orchestration together.
Check it out for yourself with these resources: