Is Legacy Technology Still Your Greatest Asset?
Who has actually made the move off of the mainframe?
The answer is almost no one, we still all have them sitting in the basement doing what they have always done.
The same thing has happened with AS400’s, racks, blades, unix and pretty much everything we ever installed (maybe with the exception of minis, these were the one generation of tech that was replaced by Unix, and we all still have fond memories of the DEC 10 , WANG or HP3000). They stay because the cost and complexity of change is just too great.
But you do get to a point where the human capital needed to keep these older systems running becomes the biggest risk. Until that point is exceeded it’s just easier to keep them around, and just connect them to newer environments where newer code is running.
Old systems become stable systems, doing what is needed without drama and with a very defined and controllable (maybe some would say comfortable) cost envelope. But they do create a weight on the business, one that slows down agility and causes us all to feel uneasy about our ability to deal with rapid changing needs.
The challenge is often that with so many generations of legacy code running on legacy systems (even if these are new versions of these old platforms), connecting everything together can become of the hardest tasks operations has to maintain. Messaging middleware has become the glue that allows all of these systems to interconnect. But once implemented these messaging middleware systems are all too often ignored from an operations perspective.
The classic monitoring tools that are used have generally failed to allow end-to-end monitoring of the entire application stack. The tools used on legacy platforms provide deep knowledge about how these environments are running, while the tools running on modern or bleeding edge platforms provide deep knowledge about how these modern platforms are running.
Today’s challenge is no longer platform centric, now the business needs to know how the user’s experience is impacted by every step it passes through. If a user experiences a slow response from their web-based service, you need to know why. With very complex business processes where hundreds of sub-applications are working together, the reason is more often than not that the interaction of many processes, each of which are working within their agreed service level, are delivering an unexpected overall outcome. The classic APM model of measuring each systems performance doesn’t allow for these more complex issues to be effectively observed.
The very first mechanical computer created by Charles Babbage suffered a similar issue. Each and every cog and wheel were made within tolerance, and yet when they were all assembled, they failed to rotate. It was an almost unsolvable issue. Today’s (nearly) infinitely more complex environments suffer from a very similar tolerance issue, and the APM way of trying to solve these issues is often no more effective than those employed by Babbage (and Lovelace).
The answer (as it always seems to be with computing) is to move to a new level of abstraction. One that allows the business processes to be considered holistically, where there is a dimension of knowledge about the way that processes flow through an application stack that is available, but often overlooked.
This is messaging middleware
Messaging middleware is used to allow all of the components of a business process to interconnect, and if you can use the configuration information within these environments along with the contents of messages you can learn a lot more about how your business is running. Combine this messaging middleware intelligence with the performance data of each system and you have everything you need to view your business from a new perspective, one that can provide end-to-end visibility of each user’s experience.
Then by using machine learning artificial intelligence, you can compare any users interaction or group of interactions (transactions) to the historical record of previous interactions to spot incredibly subtle early indicators of issues, well before they become apparent. This is how you move to the next level of business performance, by being truly predictive, solving problems before they become problems. Some people are referring to these ideas by the marketing term “AIOPs”, but there really is no formal definition of AIOPs in place yet, and so it does feel a lot like another in the long line of alphabet soup that marketing people use to justify change. That’s not to say the concept AIOPs isn’t going to be the future of operations.
AIOPs should feel like the movie Minority Report, where precognitives floating in a pool of gunk were able to predict who would commit a crime before it happened, allowing Tom Cruise to run off and legally arrest the right person before they committed murder.
If you can abstract business understanding from messaging middleware intelligence, overlaid with system performance data to provide a visualization of the entire business process irrespective of platform, infrastructure or application, you can move out of the war-room mentality and move into true root-cause analysis. Today only one company is able to provide this level of solution, and that’s Nastel Technologies.
If you feel your operations abilities are closer to Victorian Babbage thinking than Minority Report and you want to move to a predictive method of thinking, consider Nastel.
With Nastel you are able to see your users’ experiences through the entire application stack without limit. Your business uses mainframes, AS400, datacenters with blades and discrete servers, VM’s and a mix of private, public and hybrid clouds. You need to be able to follow a user’s experience through all of this, it’s no longer acceptable to see legacy systems as black boxes, you need to know exactly what is happening at every level, inside databases, inside mainframes and across managed file transfers, through blockchain environments and the physical world through IoT. This is what Nastel delivers, and it’s the secret that the world’s most successful (and largest) companies have been relying on.
Come and see what we can do for you today!