and Mesosphere’s strategic partnership enables microservices and serverless computing for modern data centers

Iron_Mesos_ALots of people have been asking about how they can use Mesosphere and together. It makes sense, because’s workload processing engine and hybrid microservices architecture are perfectly suited to take advantage of Mesosphere’s Datacenter Operating System (DC/OS), the first open and comprehensive platform for building, running and scaling modern enterprise applications. customers already love the simplicity of the serverless platform, providing a multi-cloud,  Docker-based job-processing service. Using DC/OS with, joint customers would see enhanced flexibility to develop their hybrid cloud strategy and run distributed job processing across heterogeneous environments. This would be a big win for enterprise customers who want to pair the scalability and operational ease of DC/OS with the serverless and microservices expertise of


This is why our two companies are working together to create a series of reference architectures for enterprises looking to adopt microservices at scale, for a variety of modern applications and workloads using a containerized infrastructure.

Evolution Enterprise Applications


Benefit to customers

DC/OS drastically simplifies the process of running applications, including those built around Docker containers and big data systems. It is built on a foundation of Apache Mesos, which works under the covers to aggregate up to tens of thousands of machines into a single resource pool. DC/OS services are highly available, and users can easily deploy and manage distributed services, as well as massive collections of application containers, without configuring servers. provides a hosted service that can deploy workers across public and private cloud environments. These IronWorkers are instantiated in the appropriate target clouds and execute their business logic before being destroyed.


More often than not, IronWorker jobs connect to enterprise big data services and SQL/NoSQL databases. This is where the power of DC/OS can play a key role, by allowing infrastructure developers to build distributed systems that dynamically leverage a range of underlying resources, offering the following key benefits:


  • Hybrid Microservices: Both and Mesosphere are able to support and run across public, private and hybrid cloud environements. This allows customers of all sizes from startups to large enterprises to deploy “Hybrid Microservices,” allowing flexibility of workload pipelines and back-end data services to be assembled based on performance, security, price and other constraints that customers care about.
  • Event-driven resource optimization: As IronMQ queues grow, they can set off triggers that can be used by DC/OS to expand the targeted resources.
  • Polyglot languages, meet polyglot databases: The benefits of being polyglot has been widely publicised in cloud-native environments. Combining’s support for multiple languages with Mesosphere’s ability to run a variety of big data and analytics platforms provides the greatest flexibility in the industry in a turnkey approach.
  • Containers-as-a-Service (CaaS): Specifically for, but not limited to born-in-the-cloud SaaS companies, the evolution of their platforms to microservices can be accelerated through the use of a CaaS environment. Our combined solution provides comprehensive CaaS capabilities.


A high-level overview that shows how and Mesosphere DC/OS work together, highlighting some of these benefits can be found below. For more background, you can also check out my recent post on hybrid capabilities.  HYPERLINK TO: Stack


Enabling IronWorker on Mesosphere

  1. Sign up for an account at
  2. Contact for access to the Hybrid Cluster configuration
  3. Log into
  4. Go to My Hud Dashboard
  5. Select create a new cluster
    1. Name can be any name
    2. Memory: This is the amount of memory to restrict the worker to for the workers in the cluster.
    3. Disk Space: This is the amount of disk space the container for workers will have access to when running.Clusters
  6. Select your cluster and keep track of the Cluster ID and Cluster Token as we will need these two items later.Internal Clusters
  7. Log into your DC/OS Cluster
  8. Select Services
  9. Click on Marathon
  10. Select “Open Service”Marathon UI
  11. In the Marathon UI select “Create new Application”
    1. You can either do the step by step using the UI or copy and paste the json from in JSON mode.Marathon Hourly Tasks
  12. In the New Application Dialog give the Application any name
    1. You can specify the number of CPUs to make available to each container and the amount of memory
    2. IronWorker will automatically break the memory up into workers based on the available memory. So if there is 2048 memory with 512MB workers then there would be 4 workers per instance.New Application
  13. In Docker Container
    1. Image: (This is the image that will be pulled for the container)
    2. Network: Any option should work as long as the container can connect outbound to 443 *
    3. Select “extend runtime privileges”
  14. Select Environment Variables
    1. Create three Environment Variables
      1. CLUSTER_ID=CLUSTERID (This is the value from the my clusters page)
      2. CLUSTER_TOKEN=CLUSTERTOKEN (This is the value from the my clusters page)

New Application 2

  1. Select Health Checks
    1. We’ll be creating 3 health checks
    2. Healthcheck 1
      1. COMMAND: ps -Aef | grep -q ./runner
      2. Grace Period: 300
  • Interval: 60
  1. Timeout: 20
  2. Max Consecutive Failures: 3
  1. Healthcheck 2
    1. COMMAND: ps -Aef | grep -q dockerd
    2. Grace Period: 60
  • Interval: 60
  1. Timeout: 20
  2. Max Consecutive Failures: 3
  1. Healthcheck 3
    1. COMMAND: docker ps -a
    2. Grace Period: 300
  • Interval: 60
  1. Timeout: 20
  2. Max Consecutive Failures: 3
  1. Select Optional
    1. On your mesosphere nodes you will need to do “docker login” and login with the user who has access to the IronWorker Docker image. Once you have done that you will need to gzip the .docker folder with the config.json and make sure it is available on all of the nodes.
    2. Put the tar.gz file on each node where it is accessible toDC/OS
    3. In URIs put the path to the file like: file:///etc/docker.tar.gzNew Application 3
  2. Select “Create Application”
  3. You can now click scale to have instances created



We are truly excited about this partnership and the potential it holds for turnkey “hybrid microservices” deployments. I’m looking forward to seeing how customers innovate around our combined platforms. Please share your comments and feedback below about your use cases, additional areas we should focus on, or anything else really. And stay tuned for a lot more from this partnership in the near future.


Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.