TEMBO Regroups, Preps for New Database Modernization Push
January 19, 2015 Alex Woodie
The folks at TEMBO Technology Lab knew a tough road was ahead when they set out to fundamentally transform how IBM i shops use DB2. You don’t tell IBM i pros that they’re doing it wrong with record-level I/O, and not expect to get a funny look or two. But to its credit, when TEMBO encountered resistance to its SQL mantra, the company stepped back, regrouped, and figured out another approach.
“It’s been a long hard slog,” says TEMBO CEO Marinus Van Sandwyk. “The one thing that really caught us off guard initially was we had to explain to these users why it’s necessary to move to the SQL personality of the database. We’ve been very hard at this for the past 2.5 years, to actually educate and explain to people the value they can get by moving to the SQL engine.”
The central thrust of TEMBO’s argument is that IBM i customers and IBM i ISVs will get significant benefits by moving away from the traditional DDS structures that have defined RPG and DB2 for i development for generations, and move to the newer SQL engine that IBM makes available within DB2 for i.
Doing so not only reduces the amount of application code to maintain–since triggers, referential constraints, and database validations now live in the database instead of the application code–but it also paves the way to implementing a modern Model-View-Controller (MVC) architecture. For IBM i ISVs, this means it’s just a hop, skip, and a jump to adopting a software as a service (SaaS) cloud delivery model.
Database modernization as a concept existed before TEMBO came onto the scene, and there have been many discussions, Redbooks, and articles written about how best to get IBM i shops and ISVs to adopt DB2 for i’s SQL engine. But there’s no denying that Van Sandwyk–an accomplished IBM i high available product architect and unabashed backer of the system–helped to reignite the database debate when TEMBO arrived on the scene in 2012. The combination of Van Sandwyk’s technical bonafides and his commitment to saving the IBM i platform make for a compelling story.
But TEMBO has several things working against it. First and foremost, the old mantra “If it ain’t broke, don’t fix it” looms large whenever somebody proposes to dramatically re-architect an application that has run for decades. In this age of tight IT budgets, asking IBM i shops to trust TEMBO that record-level access is holding them back takes a leap of faith. There is a certain amount of stationary momentum that is tough to overcome, but that isn’t stopping TEMBO from looking for its moment.
Secondly, the database issues that TEMBO is trying to tackle are complex and defy simple explanations. IBM i shops historically haven’t had to hire database administrators because the IBM i platform so effectively shields the programmers from database complexity. TEMBO wants to rip away that old crutch and show them how much quicker they can be without it, but IBM i pros are still clutching it tightly. If Charlton Heston were an RPG programmer, he’d have said “You can have my record-level I/O when you pry it from my cold dead hands!”
Lastly, the fact that TEMBO is based 10,000 miles from the United States doesn’t help. It’s tough for any foreign company to break into the U.S., which is the biggest single market for IBM i products and services. The fact that the South African company is promoting concepts that are foreign to so many old-school RPG programmers does not help its cause.
While TEMBO’s introduction to the US has been bumpy, Van Sandwyk isn’t giving up and will continue forward with his plan to give IBM i shops a lifeline, even if they don’t know they need it yet.
“The platform is so forgiving that people just keep on doing what they’ve done 20-25 years ago,” Van Sandwyk says. “It’s definitively not in the best interest of their organizations [to] keep on doing what they do. But until you actually show to them what it means to the business–from a lack of agility perspective. . . and also the database quality they will have–once you highlight that to them, they actually realize what it is we’ve been trying to say.”
Boiling the message down into a bite-size description has been tough for TEMBO. “One of the things we’ve been grappling with for a long time was to get the message to the level where most people, within two to three sentences, actually understanding what you’re saying,” Van Sandwyk says.
“Remember, the fundamental issue here,” he continues, “is 80 percent of the lines of code in your heritage application belongs in the database engine. I’m talking about all the database validations and relationships, which should be maintained by the database engine and not by your coding. . . . You can imagine, if you get rid of 80 percent of your metadata that you maintain by hand, what happens from an accuracy, quality, and productivity perspective.”
Keeping On Message
TEMBO’s flagship product, called Adsero Optima (AO) Foundation, automates much of the work of migrating the old DDS code into DDL and the SQL engine. The company also developed a product called AO Logic to keep the old database and the new database in sync. Van Sandwyk realized having two separate products was confusing customers, so TEMBO rolled all of AO Logic’s functionality and embedded it into AO Foundation. There will be additional products launched soon.
However, just getting the bulk of one’s IBM i application codebase running under the SQL engine is only half the battle. The migration to the SQL engine is not an end, but rather a means toward total application modernization. That has to be the goal, Van Sandwyk says, or else everything else is for naught.
“We were talking so much about the fact that people had to migrate to the SQL engine that people had the perception that that’s what we are all about,” he says. “The migration is just the first step in a true modernization. The bulk of AO is management functionality to actually mange your SQL database and databases.”
There’s very little benefit to be had just to move to the SQL engine if IBM i shops don’t take the next step and modernize their application. “It’s extremely important to understand that the first step of what we’re suggesting is the move to the DDL and then use that as a foundation for true modernization of your underlying application,” Van Sandwyk says.
The IBM i application ecosystem is not what it used to be, but there are still hundreds of small and midsize boutique software houses that develop IBM i applications for particular verticals or geographies. Getting these shops to get their highly tuned and competitive applications ready for next-generation architectures and software as a service (SaaS) delivery models is the stated goal of TEMBO.
TEMBO is targeting ISVs with the latest release of AO Foundation, which shipped in late 2014. The new database schema shadow copy capability in version 4.4 will simplify the task of replicating database schemas across hundreds or thousands of client machines. It also brings new metadata management features, including the capability to replicate field-level changes without time-consuming coding.
A New Reseller Channel
Finally, TEMBO is striking out on its own with a new business partner model. The partnership that TEMBO formed with inFORM Decisions in late 2012 ended in late 2013, and now the company has its own office in the Minneapolis area. Patrick Sheehy, who holds the title Global Modernization Champion, is heading up TEMBO’s Minneapolis office and will be establishing a new reseller channel composed of IBM business partners and consultants who are active in application modernization. The company is also looking to expand in Europe and Asia.
The U.S. plan is to get the TEMBO product into the hands of local IBM i technologists who can help their own customers. “What we found is there’s a lot of smaller niche players in the various local markets, somebody who only focuses on modernization in the Atlanta area or the Twin Cities are or greater Massachusetts area,” Van Sandwyk says. “Those are the people we will now partner with, because those are the people who have stayed abreast of the latest developments on the platform. Those are the people who are prepared to actually leverage the inherent capabilities of the platform.”
TEMBO took its first crack at the US market, and it is picking itself back up and is giving it another shot. Van Sandwyk is determined to see the modernization trend take hold of his beloved platform and is genuinely concerned of what may happen if it doesn’t.
“The more people who talk about database modernization and modernization in general, the better for the industry. That’s the fundamental message,” he says. “There’s definitely recognition globally that people have to move [to the SQL engine]. It’s as simple as that. They have to modernize, and if they don’t modernize, they will lose their applications eventually, and also the platform as a result.”