Hazelcast provides a streaming and memory-first application platform for stateful, data-intensive workloads on-prem, at the edge, or as a fully managed cloud service. The company recently announced version 5.0 of the Hazelcast Platform in which the Hazelcast IMDG and Hazelcast Jet products have been merged into a single product to not only simplify the development of applications using Hazelcast but also put more focus on developing real-time applications that will engage with end users in new ways. Hazelcast CEO Kelly Herrell shared more information about the new release as well as the company’s future road map.
Earlier this year, Hazelcast announced Hazelcast Platform 5.0. What is the significance of the name change and how does it relate to the IMDG and Jet products?
Businesses no longer want to act on what happened yesterday; they want to act on what’s happening right now. That requires a new, tightly integrated combination of technologies. Gartner captured this in a recent report that talks about the “collision” of several technologies, including streaming and in-memory computing. Their report directly aligns with what we’ve been seeing from our customers and, as a result, we believe we’re one of the first, if not the first, to combine these technologies to help our customers deliver real-time applications.
At the product level, the term “stronger than the sum of its parts” applies here. Hazelcast Platform integrates what were two separate products: Hazelcast IMDG, an enterprise-class in-memory data store that accelerates the storage, retrieval, and modification of data, and Hazelcast Jet, a high-performance, low-latency distributed data stream processing engine. This powerful combination gives enterprises a single and simplified platform for transactional, operational and analytical workloads.
What does this mean for users of the previous products?
From an upgrade path, all current customers can upgrade to the new Platform to take advantage of our in-memory data store and stream processing capabilities. They will be able to add new capabilities to their existing applications at no additional cost.
When it comes to the upgrade path itself, we have been focused on maintaining compatibility with existing users. For example, Hazelcast Platform 5.0 is fully compatible with Hazelcast IMDG 4.x. We are also providing a zero-downtime upgrade path for Hazelcast IMDG 3.x users, although some application code changes are required. For existing Jet users, the upgrade will require stopping and resuming jobs, as has been the case when upgrading between Jet releases.
Can you highlight three of the most important new features and how they better address changing customer requirements?
The biggest feature is really the focus of the whole platform: enabling true real-time applications. To do so, we’re making it easier and faster for organizations to process streaming data as it’s generated and merge it with historical context?data stored in databases or elsewhere?to deliver real-time applications that can capture value that is otherwise outside their technology reach. This provides new, timely and truly meaningful opportunities for customers to engage with applications. That’s a huge competitive advantage.
But three other important new features include expanded SQL support with many new expressions, new persistent storage functionality that adds critical resilience, and increased storage capacity. When it comes to the latter, Hazelcast Platform supports 200GB per member, effectively doubling the maximum storage capacity on the same hardware. This increased density, combined with the linear scaling of the number of members, helps companies face exploding storage requirements while maintaining extremely high performance
What is the hurdle that you see among organizations trying to get more from their data?
One of the biggest challenges organizations are facing is merging real-time streaming data and data at rest. In fact, a recent global survey of IT decision makers showed that nearly four of five professionals are having difficulty with this issue at some level. Organizations have had an enormous challenge getting their hands around big data; now they are realizing the potential of merging the volumes of data they have stored with the massive amounts of data pouring in from servers, mobile systems, IoT devices?you name it. The value in that data is highest the moment it is born; the challenge—and it’s a big one—is harnessing that data in real time and integrating it with data at rest in different systems.
What do they need to change in terms of technology or processes?
It’s less about changing technologies, and more about processes and mindsets. The technology landscape for data solutions has become increasingly complex and fragmented, and enterprises are drowning in decentralized data. As a result, internal teams have become more siloed because you need experts in different areas. This complexity is the enemy of innovating quickly in a hypercompetitive world.
Hazelcast rises above this complexity by providing a single execution environment that connects to all sources of data, resolving the collision described in the Gartner report. Its immense processing speed enables a simple and unified way for developers to create innovative real-time applications and focus on the customer experience and outcome.
What do you expect that companies would be able to accomplish if they could improve in these areas?
By creating real-time businesses, our customers are gaining accelerated revenue growth, lowered risk profiles, and increased cost efficiencies. For example, BNP Paribas created an application that enables real-time customized loan offerings—versus sending an offer days after it was needed—that grew a product line’s business by 400%. We’ve seen similar impacts for other customers in the areas of fraud detection, recommendation engines, alerting, and supply chain and production management.
What’s next on the Hazelcast road map?
Hazelcast delivers an open and integrated platform with all critical capabilities required to build a real-time solution for your business. At a high level, our roadmap includes providing our customers with the best product experience for building real-time solutions; democratizing streaming analytics by improved pre-integrations with commonly used messaging systems and data sources; and making SQL the lingua franca for accessing and analyzing all data?data-at-rest and data-in-motion.