Interesting post by Chris O’Malley, the CEO of Compuware.
What’s Wrong with the Mainframe?
- The mainframe is not too expensive. This common myth is based on decades-old denial about the true cost of distributed infrastructure. Yes, mainframes require a high capital outlay when compared to commodity PC servers. But keeping that commodity infrastructure up and running has proven to be one of the most expensive and fruitless undertakings in history. In fact, TCO for distributed computing make capital IT costs look like pocket change.
To virtualize their distributed server environments, for example, enterprises are enriching a single software company—VMware—to the tune of $6 billion annually. In stark contrast, the entire mainframe ops tools market is just $10 billion for all vendors combined. That’s why Rubin Worldwide found that 1) over the past five years, costs at server-intensive IT shops have risen 65% more than at mainframe-intensive IT shops, and 2) mainframe-intensive companies earn 28% more per dollar of IT infrastructure than server-intensive companies.
- The mainframe is not a security problem. Contrary to certain recent inaccurate reports, the mainframe is in no way a threat to any organization’s information security. Just the opposite is true. None of the many recent high-profile hacks (Sony, Target, etc.) have anything to do with penetrating mainframe defenses. On the contrary, they are almost all the result of poor security practices associated with distributed networks. The mainframe remains vastly more secure than distributed systems when it comes to intrusion exploits, malware, and every other category of security threat.
- The mainframe is not a migration problem. Despite its technical and economic superiority to distributed platforms, a surprising number of industry voices still contextualize the mainframe as a “legacy” platform from which enterprises need to migrate their core applications if they are to succeed in the digital economy.
This makes no sense. First of all, why would any organization migrate its most critical applications from a supremely reliable, secure, scalable and secure platform to a relatively risky and expensive one? And why would any CIO allocate limited resources to a low- or negative-ROI migration project when so many other urgent imperatives clamor for his or her limited IT resources?
The answer is that there is no reason. That’s why analysts like Gartner are reporting minimal migration activity—and why 88% of CIOs assert that their mainframes will run existing and even net new workloads for at least another decade.
But if these are not the mainframe’s problems, what is?
The Apathy Virus
The mainframe, it turns out, is vulnerable to a virus. But this virus is not digital malware. It’s a cultural pathology: Apathy.
Apathy has infected mainframe culture in several ways. One is the complacency that comes with success. Mainframe teams have done a nearly flawless job of maintaining the availability and integrity of core business applications and data, decade after decade. So it’s natural that they would rest on those laurels—especially since the hard work of innovation has been largely left to their peers in the distributed, web and mobile spheres.
The illogic of platform migration has also lulled mainframe teams into an apathetic mindset. Mainframe applications represent irreplaceable and highly valued intellectual property. And they aren’t going anywhere. So mainframe teams have a certain level of job security that insulates them from the intense pressures that drive innovation in other areas of IT.
Mainframe leadership has also suffered from the frog-in-slowly-boiling-water syndrome. Unlike Y2K, the looming loss of veteran mainframe professionals does not present a hard, sudden deadline. So CIOs have not felt immediate pressure to recruit their next generation of mainframe developers and operators. One recent survey revealed that as many as 39% of CIOs do not have any formal plans in place to refresh their mainframe staffs. This is obviously not a viable position as a generational shift takes place in the IT workforce.
Symptoms of this apathy include:
- Outdated approaches to application development, testing and maintenance that are insufficiently responsive to changing business requirements and insufficiently supportive of business innovation
- Failure to capitalize on major new mainframe opportunities such as IBM’s specialty Java and Linux engines
- Organizational insularity that absents the mainframe from the IT mainstream—diminishing its engagement with the intense innovation taking place in distributed, web, mobile, cloud, social and IoT computing
- Insufficiently aggressive on-boarding of top Millennial talent
- Excessive emphasis on cost-cutting at the expense of high-upside investment
Mainframe teams that continue to succumb to this Apathy virus are—without realizing it—fulfilling the prophecies of the platform’s demise. But that demise is far from unavoidable. And the alternative is extremely compelling for both IT and the business.
Transforming the culture
IT leaders can reverse the effects of mainframe apathy by transforming mainframe culture. Concrete steps in such a transformation include:
Adoption of Agile Methodologies and DevOps
Most mainframe users still operate under a “waterfall” model. They also tend to take their time about promoting new code into the production environment. This is unacceptable given the pace of change in business today. Only by adopting Agile – and making the mainframe more conducive to Agile
operations – can mainframe shops keep code tightly aligned with relentlessly evolving market demands. And only with more advanced DevOps models can businesses achieve greater agility, through accelerated code promotion.
Mainframe code, data and hardware increasingly act as back-end support for multi-tier applications with mobile and/or web front-ends. So mainframe teams must work collaboratively with DevOps teams across other platforms. This collaboration requires changes in both culture and processes.
IT can’t deliver a more vibrant and business-relevant mainframe by settling for “good enough” tools simply because they’re available at a discount. Instead, mainframe owners must invest in best-in-class technology that enables them to get maximum productivity from their staffs and maximum business value from their mainframe code and MSUs.
Prioritizing the Talent Search
Given the generational shift taking place in the mainframe workforce, it is essential for IT leaders to prioritize recruitment of Millennial talent. This recruitment requires more than upping salaries and benefits. It also requires offering meaningful projects that relate to customer experience, metrics-based decision-making and other hot areas of interest. Recruitment and retention of top Millennial talent will also require workspace tooling that doesn’t have a 1990’s look-and-feel.
All these anti-Apathy measures should be grounded in the recognition that mainframe application code represents irreplaceable, high-value corporate intellectual property developed over multiple decades. It is neither technically nor economically advisable to migrate this IP off the mainframe.
Mainframe revitalization thus offers companies a unique opportunity to improve business performance by more aggressively leveraging mainframe-resident IP—while also improving the economics of IT by moving net new workloads to their most efficient, scalable and secure computing platform.
Chris O’Malley is CEO of Compuware.