IT Management

Attention big-iron warriors! The Dell Tech Force Alliance is out to crush the leviathans of big iron! Sound familiar? It’s the same song from the ’90s when mainframe bashing was in vogue and countless pundits yet again mispredicted the death of the mainframe. Also, please note the subtle difference; those singing the song this time around are attempting to tag high-end Unix servers (HP, Sun, and IBM) with the “big iron” label. This is big irony because the Dell alliance is using messaging similar to what HP and Sun used years ago to lure customers away from S/390 and onto high-end Unix servers.

Check out Dell’s video at www.delltechforce.com and see if you can guess what age group is being targeted. According to the script, “Now there is a powerful alliance of heroes, a new force that will change the world.” Enter force leader Michael Dell and his band of billionaires: Bill Gates (Microsoft) and Larry Ellison (Oracle), plus the chief execs from Intel, EMC, and Red Hat. This sextuplet forms the Dell Tech Force Alliance who will come to the rescue whenever “Big iron is attempting to lock customers into high-cost, low-return solutions.” Their answer is to offer scalable solutions, pointing out that “with standards, customers get better value, performance, reliability, and flexibility.”

Like me, are you also wondering how a loosely knit alliance, that can unravel as quickly as economic conditions can change, will crush a sophisticated environment that has evolved from several decades of closely coordinated hardware and software development and represents a gazillion dollars of customer investment? The alliance certainly has their work cut out for them! zSeries sales posted positive revenue growth for the second consecutive year; IBM is presently positioning the zSeries to function as the rock-solid, bulletproof, worm-free, virus-free, intrusion- free, self-optimizing, self-healing hub of a heterogeneous computing utility that will be used to consolidate runaway server farms and provide economical means of offering shared services. When the six-company alliance comes to the rescue, ask who is going to integrate the final solution. Most likely, that would be you!

Let’s not forget the scores of technology fashion victims that have believed the myth that scalability and running mixed workloads is as simple as adding another server. To offer a legitimate mainframe alternative, the alliance must solve a number of scalability issues when resource serialization, processor contention, path-length explosion, and cache coherency rear their ugly heads. Fred McCallum, former director of IT for the Canadian government, once told me he countered the scattered computing nonsense with this simple question: “Would you prefer to pull a plow with an ox or with a thousand chickens?” Alas, farmers may have common sense about proper use of chickens but runaway server farms suggest common sense isn’t all that common in the IT industry. Let me be perfectly clear, I’m not saying the mainframe is always the best platform for any application. I’m saying the zSeries mainframes my clients manage are clearly superior to Dell servers in a number of key areas, including total cost of ownership, scalability, reliability, availability, serviceabilty, logical partitioning, security, resource sharing, workload management, systems management, automation, and data movement, to name a few.

By now you may be asking, “But Bill, how can you be so bold as to challenge the claims of three famous and successful billionaires?” I’ve checked too many alleged mainframe downsizing success stories only to find people who had been fantasizing instead. For example, I followed up on press reports about a cutting-edge bank in the U.K. whose CIO had convinced his top brass they would leapfrog their competition by dumping the mainframe and moving to 2,000-plus Wintel servers. I found a bank whose IT was in disarray, the CIO had long since abandoned ship, and eventually a competitive bank acquired them!

I still contend that every tool has its place, so please send me legitimate big-iron-crushing stories so I can showcase them on my Website. I’ll also write about legitimate mainframe vulnerabilities (e.g., ISV software costs, skills shortages, and a number of application development issues) in the next issue of z/Journal. It’s important to understand the strengths and weaknesses of each platform so we can choose the best tool for the job.

Which is the best platform? It depends on what you’re trying to do. If we first understand business requirements, our chances of choosing the right technology improve dramatically.

Grace and peace.