The IBM MQ AWS Partner Solution (MQPS) is a deployment of an IBM MQ Native HA queue manager running on the AWS Elastic Kubernetes Service (EKS) platform. Whether you’re an Integration architect or system administrator who wants to quickly evaluate the viability of running a highly available IBM MQ system on an AWS container platform, or a developer looking to get started quickly, this solution allows you to easily deploy an IBM MQ queue manager into your AWS environment.
The new MQPS includes the innovative Native HA container-based solution that provides native replication and fail over in seconds. It also adds the ability to select the latest build of MQ so you can benefit from the newest features of MQ.
Visit this AWS page to read more about the benefits of Partner Solutions and for more developer-focused journeys for MQ on AWS, see: https://developer.ibm.com/components/ibm-mq/learningpaths/
The platform
The MQ container-based solution will be deployed to an AWS EKS cluster. EKS has been chosen as the platform to provide more of an AWS-centric solution. To manage the deployment of MQ and the queue manager, the EKS solution uses Helm charts. If you would like to look at the sample Helm charts, they are available here: https://github.com/ibm-messaging/mq-helm. Note that the management of the MQ deployment on EKS differs from a deployment using the Red Hat OpenShift platform. Typically OpenShift deployments will use the MQ Operator, presently the Operator is only available for OpenShift.
The automated deployment can use either an existing AWS Virtual Private Cloud (VPC) or create a new one, the deployment instructions have both options. Unlike the earlier solution, the MQPS uses EBS storage instead of EFS, this is because the queue manager itself will manage data replication.
The IBM MQ Partner Solution is a reference deployment of an IBM MQ Native HA queue manager running on the AWS Elastic Kubernetes Service (EKS) platform.
Important Notes:
The Partner solution uses the IBM MQ Advanced for Developers container image and uses sample Helm Charts for deployment - both of which are not supported by IBM.
For production deployments of MQ on a non-OpenShift container platform, you should build your own container and host it in a container registry. Full details on how to build a supported MQ container that can be used with a production license entitlement can be found here:
https://www.ibm.com/docs/en/ibm-mq/9.4?topic=integration-planning
Get Started Today!
The first thing you’ll need is an AWS account. You can sign up for an AWS Free Tier account on the AWS homepage here: https://aws.amazon.com/free
When you have created an AWS account, go to the MQ AWS Partner Solution page here: https://aws.amazon.com/solutions/partners/ibm-mq
Click on the “view deployment guide”, read the instructions and then click on one of the deployment options: either a new VPC or an existing VPC. A new user will probably want to choose the new VPC option. If you are not already logged in, you will need to log in to your AWS account before proceeding.
Once you have logged in, the automated deployment will present you with a few self-explanatory options, complete and submit the form and you will have a queue manager deployed into an EKS cluster in minutes. The MQPS page above has detailed instructions on how to connect to and test your queue manager.
Summary
The MQPS is a quick and easy way to get a queue manager running on a Kubernetes platform, you can use it to explore the remarkable capabilities of a Native HA queue manager, and when you are ready to take the next step of deploying to your enterprise you will have the confidence to know you are choosing a best-in-class solution that you can rely on to maintain the integrity of your integrated systems.
What kind of queue manager will the solution deploy?
The MQPS creates an IBM MQ Native HA deployment using the latest release of MQ Advanced for Developers. While this uses a developer edition, the actual queue manager deployment still represents a highly available replicated queue manager that has availability across 3 AWS availability zones.
MQ Native HA is IBM’s very latest container-ready, highly available queue manager solution. MQ Advanced for Developers has a free license that allows you to use MQ in a single developer deployment, full details can be found here: https://developer.ibm.com/articles/mq-downloads , in short – you can use MQ Advanced for Developers for your own single-user testing but not to test things like application end-to-end testing or performance testing. For more information on MQ Native HA queue managers I recommend you watch this introductory video: https://www.youtube.com/watch?v=LHjfhNR6tKE (2min).
Important notes:
Although the MQPS will deploy a highly available replicated queue manager that you can use as a reference to create a production deployment. There are several reasons why it is not recommended to use this as delivered in production:
The MQPS relies on a separate EKS Partner Solution to stand up an EKS cluster. This can be quite an expensive way of deploying as it is for a single queue manager for evaluation and test purposes. A more typical production MQ container-based deployment should use an existing fully-managed Kubernetes cluster. The many tasks necessary to create the EKS cluster and deploy the queue manager have been automated as part of the MQPS but it still requires user input at deploy-time. Best-practice production deployments should be fully automated using a repeatable configuration or infrastructure as code methodology with production-quality controls and governance. Most of the user input is related to the EKS cluster and the VPC, so this is not a major concern for further automation.
The MQPS uses MQ sample Helm charts which are not supported. It is recommended that you fork the repository and maintain the Helm charts yourself in your own production code repository.
The container used by the sample Helm charts is the developer edition of MQ. To deploy MQ on a non-OpenShift container platform, you should build your own container and host it in a container repository. Full details on how to build a supported MQ container that can be used with a production license entitlement can be found here: https://www.ibm.com/docs/en/ibm-mq/9.4?topic=integration-planning
#IBMMQ
#Spotlight
#automation-featured-area-1