Syncsort CEO Discusses Vision Deal, Product Plans
August 28, 2017 Alex Woodie
Josh Rogers isn’t one to get overly animated, at least in an over-the-top “Ra Ra” kind of way. But in a conversation with IT Jungle last week, there was no disputing the Syncsort CEO’s excitement at closing the acquisition of Vision Solutions and the prospect of delivering big data integration products to a new pool of IBM i customers.
To quickly recap: In early July, Centerbridge Partners, a large equity capital firm, acquired controlling stakes in both Vision Solutions and Syncsort, and announced plans to merge the two companies under the Syncsort brand. Two weeks ago, the vendors formally announced the deal had been completed. Rogers, who was previously the CEO of Syncsort, continues as chief executive of the combined company.
For Rogers, the 4,000 or so IBM i customers that Vision brings to Syncsort – which developed ETL and data integration software for IBM System z mainframes, Hadoop, and Splunk before beginning to diversify itself through acquisitions about a year ago – represents a familiar user base.
“The IBM i platform is a terrific platform for running transactional workloads,” Rogers says. “It’s highly optimized for that and has been for years. It’s highly available and reliable, and also easy to manage. We really like that market. And when you think about the work we’ve done in the big data space and the opportunity that you and I have talked about around ‘Big Iron to Big Data,’ these exact same things hold for the IBM i platform as they do for the mainframe platform.”
Syncsort will continue to develop and sell Vision’s high availability software. Vision, of course, emerged as the dominant vendor in the rough-and-tumble high availability software market for midrange servers. But the name “Vision Solutions” has come to an end, although the MIMIX and iTera brands will continue to be used.
“We are 100 percent committed to those product lines and will continue to enhance those product lines,” he says. But the bigger opportunity, Rogers says, is developing and selling big data integration tools to IBM i shops to allow them to embark upon data analytics projects using next-gen architectures, such as Apache Hadoop and Splunk.
The idea here involves leveraging the existing IBM i investment that Vision made in data replication software – in particular the MIMIX Share product – and hook that up to DMX-h, the data integration product that Syncsort developed to move and transform data from mainframes into Hadoop.
There are also plans to expose IBM i data via IronStream, which Syncsort originally developed to offload log data from mainframes into Splunk’s proprietary NoSQL data store and analysis environment.
“Just like we’ve seen a lot of success in developing new capabilities for mainframe environments, we think there’s the same opportunity in the IBM i space,” Rogers says. “We’re very excited to work with the Vision team to think about new solutions to bring to market. There’s a pretty good backlog there based on what they’ve been working on.”
Anecdotal evidence suggests that traditionally conservative IBM i shops aren’t making huge investments in big data analytics – at least as far as the IBM i platform itself is concerned. The platform and the applications that run on it are decidedly tilted toward the transactional end of the workload spectrum, as opposed to the analytical end. Sure, line of business groups may work with the IBM i professionals in the IT department to ETL data out of the IBM i server into data warehouses of various flavor. But IBM i pros themselves have largely kept their distance from the frenzy, hype, and hoopla that has accompanied the big data analytics explosion.
But Rogers says there’s plenty of interest among IBM i-using companies to utilize next-gen big data platforms, such as Hadoop. In fact, Rogers says he’s seeing the same level of data analytic investments and technological sophistication in larger IBM i shops as he’s seen in the mainframe shops that Syncsort works with. Syncsort counts about half the entire IBM mainframe installed base as customers.
“Some of these [IBM i] companies are the largest companies in the world,” he says. “If you look at the spread of revenue across industries, there’s a little more that comes out of distribution and manufacturing and retail than if you look at Syncsort. But that doesn’t mean that those organizations aren’t rapidly investing in embracing big data and advanced analytics. We’re excited to be able to offer sophisticated and best-in-class integration options, that will not only address IBM i, but will give us better exposure to a broader set of industries.”
In addition to DMX-h and IronStream, Syncsort has some cards up its sleeve in the form other projects on the roadmap, including integration with the data quality software that Syncsort obtained with its Trillium acquisition in the fall of 2016. The firm will have more details to share later this fall, Roger says.
“Our goal is to continue to integrate the various capabilities that have been acquired over last year,” he continues, “and continue to solve this ‘Big Iron to Big Data’ challenge, which is giving you best in class connectivity and access to these legacy data platforms, allowing you to move to that at the speed of business, whether that’s large volumes in batch or smaller volumes in real time, and allow you to improve your trust in that data by improving the quality of the data leveraging the Trillium technology.”
Hadoop and Splunk are popular distributed storage and processing platforms for big data sets today. As popular as these platforms are today – and as largely foreign they are to the folks running systems in traditional IBM i and mainframe shops – there are other big data platforms on the horizon, and Syncsort hopes to be the trusted envoy introducing IBM shops into that world and guiding their path through it.
“You can imagine a scenario where I have a solution where I can connect the mainframe or IBM i to a wide variety of other data repositories,” Rogers says. “Those platforms [IBM i and mainframe] are generating all the core reference data for these large enterprises. It’s all the transactions that all their customers are doing. It’s critical as these firms pursue a next-generation analytic strategy that those critical data assets are available for these next-generation repositories.”
Going forward, it will be interesting to see how Syncsort’s relationship with IBM itself changes. The Pearl River, New York company already had a formal partnership with the Armonk, New York tech giant for its mainframe work, and now Syncsort is inserting itself front-and-center as a major player in the IBM i world, too.
As far as the Hadoop story goes, IBM has ended development of its own distribution and standardized on the Hortonworks Data Platform (HDP) from Hortonworks. HDP is the only distribution that’s been certified to run on IBM’s Power processor (for what it’s worth, HDP is also the only Hadoop distribution certified to run directly on Windows, thanks to Hortonworks partnership with Microsoft). Syncsort also has a partnership with Hortonworks and works closely with the Santa Clara, California firm to ensure that its software runs on new releases of HDP.
Where all this goes is anybody’s guess, but it’s interesting to see how the pieces and players seem to be aligning relative to the IBM i installed base. Getting meaningful results from big data analytic projects is not an easy thing, and the tools are immature in many ways. But the field is rapidly maturing, and as the tools get better, even the smaller IBM i shops will feel the pressure to use them.
When those IBM i shops are ready to begin executing their big data plans, Syncsort hopes to be there with tools that are ready to go to work.