Everyone wants or needs to build and manage real-time systems these days. With the move to the edge, and growing reliance on artificial intelligence and machine learning, the power and potential of real-time computing has come to the fore. This calls for greater observability, enhanced leveraging of data, and perhaps most importantly, a well-prepared organization. Is everybody ready for this?
First, there’s a need to understand what real-time is — definitions and perceptions of what constitutes real time have been all over the map. “The notion of what real-time means varies depending on who you’re talking to and the specific domain,” says Joseph George, VP of product management with BMC Software. “While real-time in mission-critical environments requires data to be processed within milliseconds or even microseconds, real-time in the context of digital transformation and meeting customer expectations to deliver online services and information may have different requirements.”
Its applications are broad. “Any application requiring instant changes to features based on external factors like user behavior, security or bugs could require real-time capabilities,” says Nick Rendall, product marketing manager of CloudBees. “For example, in a SaaS application where users are able to trial and purchase new features instantly — provisioning. Or in more advanced organizations, if a bug or security breach is detected, the ability to shut off the feature in question in real-time without redeploying becomes very important and is part of a modern DevSecOps program. These examples would be equally relevant to B2B or B2C applications.”
Real-time technology also plays a role in enterprise applications such as logistics, shipping, inventory or products. “During the pandemic, tracking shipments from international sources became more complicated,” says J. Todd Jennings, CEO of Nexterus Technologies. “Applications with real-time technology should be very sensitive regarding shipments that need to be coordinated with the launch of sales and marketing efforts.”
Real-time technologies may see applications as the Internet of Things proliferates, “With the advancement of IoT, customers not only can track their shipments, they can request specific information about the conditions the shipment experiences,” says Jennings. “For example, the Pfizer coronavirus vaccine needs to be stored at ultra-low temperatures. Real-time logistics technology allows the temperature of the shipment to be measured and tracked throughout its entire journey. The customer can monitor the progress to ensure quality control. Shipping devices can also track whether a container has been opened or fallen over during shipment. All that information, with the use of IoT, can now be tracked.”
The customer experience is also being shaped by real-time technologies. “Customers increasingly expect a transcendent customer experience that gives them what they need, when and where they want it, tailored to their preferences,” says George. “Companies need to be able to deliver information to customers as close to real-time as possible and most importantly meet customer expectations. Clearly, customer expectations have changed.”
What does it take to build a sustainable real-time enterprise? Industry experts provide the following words of advice:
Build for the enterprise. Commitment to developing and supporting the technology on an enterprise scale is another requirement. Many companies “have not updated their development and release practices to match the changes and insights provided to them by these new technologies,” says Rendall. “And many development teams do not have the bandwidth to experiment with new features at the level they would like to.”
Focus on observability. There’s a need to understand what’s going on beneath the surface as machines make the hard decisions. With the rising emphasis on real time, observability has become a key term in the IT space. “With increased complexity and volume of data, enterprises must shift from a monitoring mindset to observability and actionability in order to provide more real-time insights and support autonomous digital enterprises,” says George. “Monitoring was all about alerting based on metrics data which notifies you of what the problem is. Getting from the ‘what’ to meaningful and actionable insights involves IT operations responding to the status of an alert, analyzing issues, accessing multiple systems to confirm compliance and service-level objectives, and creating and invoking remediation actions.”
Observability is a broad enterprise challenge, as it “involves collecting a wide-range of data including metrics, events, logs, and topology to provide the ‘why’ when something goes wrong,” George says. “Actionability goes one step further to look at how you can respond to a failure beyond what failed and why it failed. It looks at what you can do about it in the moment to remediate it or, using advanced insights, get ahead of it and take preemptive action to prevent it.”
Focus on processes. “The tools need to be in place to enable real-time changes and tracking,” says Rendall. “This means mature, automated release processes and analytics that let you create, release, measure and react to data provided by real-time changes as quickly as possible. And you need to be able to do it in a way that the feedback loop is strong and consistent across customers and teams. This starts with taking a holistic approach to CI, CD, feature flags and the common analytics engine used to measure across all of these processes.”
Look to artificial intelligence and automation. AIOps and artificially intelligent service management (AISM) strategies can play a role here. “The ability to discover, monitor, service, remediate, and optimize the IT landscape enables customers to fully capitalize on data across the enterprise and make informed decisions based on real-time insights,” George says. For example, “applications where real-time is required in mission-critical environments are those that need to consume and analyze data and make life-impacting decisions without delay. Take autonomous vehicles for example, where it takes a matter of milliseconds to process data from sensors before a vehicle has to take action. Any network delays can lead to undesired consequences, so data processing and decision-making is increasingly happening at the edge.”
Take a team approach. “This is partially a cultural shift, but also one that would occur when looking at resourcing and what developers are working,” says Rendall. “The entire development organization cannot be on standby to react to real-time insights, but teams should be.” This will vary team by team, “A marketing or product team performing a real-time experiment is completely different from a sales or customer success manager wanting to provision or sunset functionality in one of their accounts, yet both require real-time capabilities,” Rendall says. “Establishing a baseline across teams for what is real-time in the organization, how it can help in their role, and what the limitations or considerations of their capabilities is key. For example, a marketing team wanting to experiment will pull resources away from a development team in certain situations, and so that SLA would need to be established upfront.”
Work from with the end in mind. “What attributes are most important to the customer?” Jennings advises asking. “From there we can identify data points, identify the workflow steps, visualize the process, and track the project. Using data collected from every step of that process we can build a dashboard to monitor the steps. By having the map and defining the workflow and data, we can extrapolate that out into other capabilities.”
Pay attention to your data. It’s important “to have a full understanding of your organization’s data and its current capabilities to visualize, secure and harness it to inform actionable insights,” says George. “Based on this, organizations must assess where they are in relation to their business goals and objectives and map a course that will guide all facets of the business both internally and externally. When assessing where they are on this journey, it is vital to identify which systems and datasets are the most critical for what they seek to achieve, as these are the systems that must be optimized and automated.”
