MultiValue Tools

When I joined Rocket Software in 2017 as the MultiValue Tools Senior Product Manager, I had no idea that I would have the chance to collaborate with some our most innovative customers, software developers and the marketing team within Rocket Software, to bring to market a truly new and revolutionary gateway to the MV Application Platform (UniVerse, UniData) – the MultiValue Integration Server (MVIS). 

MVIS easily extends core business logic currently hidden within your application, enabling you to build and interface with modern software paradigms. The journey to develop the MultiValue Integration Server was initiated from a desire to modernize our product offerings and to attract new talent into MV. We acted by building a team of software engineers with a natural drive to stay current with constantly evolving software paradigms and tools. We also recruited some of our most forward-thinking and committed customers who wanted to move to the cloud, participate in the API economy, and maintain 24/7/365 business continuity.

Customers asked for a solution that would allow them to modernize their trusted, business critical applications. Recruiting new talent was another problem since BASIC is not actively taught in colleges and universities. With this input, the team established the following design tenets for our solution:

• APIs first
• cloud ready
• incorporate a low code/ no code development platform (LCDP)

LCDP is vital to help recruit recent graduates, who could have an immediate impact on the customers' modernization initiatives, without any knowledge of MultiValue. Two upcoming blog posts will demonstrate how Andrew Gorovoy, our summer intern, built a beautiful front end and mobile application using Flutter (via MVIS), in a matter of a week without any knowledge of MultiValue. Definitely a "Must Read" post.

With minimal guidance, within a minute Andrew was able to create a RESTful service within the MVIS Admin Console which was designed with ease of use in mind. Since MVIS is also integrated with Swagger/OpenAPI 2.0, the newly created RESTful service also included the Swagger definition. Any developer, external or internal can then exercise the Swagger definition through the client. Additionally, if a young developer wants to create a client or server mockup based on that service, the Swagger Editor can generate it in a language that matches the application for testing. If you're not really into using the Admin Console, no worries as MVIS offers a set of Admin APIs that you can use to create RESTful services, and control and manage MVIS as well.  Admin API definitions are also available via Swagger, making it really easy for your DevOps admins to use for Continuous Integration/Continuous Delivery operations.

We also wanted to break the existing modus operandi. What I'm coyly referring to here is our waterfall way of development, that I'm now happy to say is a thing of the past. The MVIS team we put together is the first team within Rocket MV that swallowed the red pill, fully embraced Agile and has never looked back. This move will allow us to deliver builds and release faster to the market. We definitely surprised our beta customers by delivering builds at the end of every sprint – builds that incorporated their feedback from a session or two ago. One of the tenets of Agile is to learn and fail fast, allowing us to truly experiment. Customers were very involved as a lot of their feedback has been incorporated into the product. For example, as I mentioned, the product was built with the cloud in mind, so we assumed that customers would utilize the cloud platform's native monitoring and logging systems (Azure – Application Insights, AWS – CloudWatch) to monitor application health. We quickly received feedback that our assumption needed revision since customers also wanted to use their existing monitoring and logging platforms such as Datadog, Splunk, or Prometheus. Traditionally we would have built APIs for each platform to export data into, but per Agile, the team took on a SPIKE or research task to find a better way. Agile sprints include research time, and we used that time well and came up with a better way – Fluentd. Fluentd integration empowers customers to export: the granular per API, aggregate, or performance statistics to 15 different monitoring and logging solutions. An added benefit, Fluentd also communicates seamlessly with many notification systems including Slack and Twilio.

Many of our customers expressed interest in the cloud, either from a lift and shift strategy perspective, as a cost saving move, or for building native cloud apps and using some of the powerful, pre-built services offered on the platforms. While some of our customers are very forward thinking and fully embrace the cloud, many are early explorers and want to test the waters a bit before jumping in. So, our team built MVIS with very flexible deployment and connectivity options, to cater to our diverse customer base. You can deploy MVIS right on top of the MV Application Platform. Or you can deploy MVIS outside of the Platform in the middle tier. You can also deploy MVIS in the cloud as a container and connect across servers in the cloud or on-prem. We've made deployment so flexible, some Yoga Yogi's might be jealous.

Today's end users expect systems to be up and running not just 24/7, but 24/7/365. You no longer have the luxury of taking down your servers to perform maintenance and upgrades and with MVIS, you don't need to. MVIS supports blue-green deployments, to help you quickly deliver a newer application to the end user, without pausing traffic. It automatically drains traffic from your old app server (blue), over to the newer updated server (green) without impacting your end users During peak demand, such as Black Friday when traffic can spike, MVIS can scale horizontally to absorb some of that load and provide a level of resiliency.

We've built so many features into this product that I can espouse for hours on end. You have to experience the power of it to believe it, from its simple installation and setup to its ability to natively run in the cloud, we put a lot of thought and effort into designing MVIS to ensure that it is the best way to access your business logic and expose it to bright young minds to modernize your applications. If you're wondering if this is as good as its going to get, the answer is "no". There's a healthy roadmap for MVIS that you can learn about during our monthly MV Roadmap livestream webinars. It plays a critical role in our MV DevOps strategy as well.

So, I am about to blow the whistle on something we've been cooking up in our innovation lab, the MV DevOps Experience. When we started collecting thoughts from our customers on why they wanted to move to the cloud, what they were expecting from the cloud, and what some of their current challenges were, we came to the realization that we have a real opportunity to simplify the whole process of deployment, provisioning, and management of MV servers and applications. The MV DevOps Experience quickly evolved from the back of a napkin concept to something we've been actively working on and hope to have a minimum valuable product (MVP) quite soon. Quintessentially, the MV DevOps Experience will allow you to lift and shift your existing on-prem solution directly to the cloud provider of your choice. Whether it be AWS, Azure, or a Virtual Cloud Provider (VPC), you'll be able to provision an instance in the cloud, manage it, and copy any application through a discovery process, publish it to GitHub, and deploy it into the newly provisioned instance. This is going to make your development, QA, integration, or even your production environments a breeze to setup, accelerating your team's efficiency to deliver more than ever before. In short, the Rocket Software R&D team is committed to delivering more!

This post was previously shared on the Rocket Blog.