Anthony Malakian: Learning Lessons

When it comes to disaster recovery, sometimes it can feel like repeatedly running one’s head into a wall. Last year, AIG ran a major disaster recovery test, looking for problems that could arise if the worst were to happen. AIG CTO Mary Kotch and her team couldn’t figure out why they weren’t able to bring up certain environments. After hours of brainstorming, they decided to start from scratch and reassess all of the environments. This time, though, they mixed up the teams, ensuring that everyone wasn’t speaking only to the same team members. Four hours later, the problem was solved.

A Story from the Past
My father worked in IT for four decades building datacenters. Kotch’s story about AIG’s IT challenges reminded me of a story my dad told me a number of years ago. Back in the mid-1970s, he worked for insurance company Equitable Life, which is now AXA Equitable.

“Data at that time needed to be chronologically organized,” my dad said. “If it wasn’t, it became impossible to avoid corrupting files. Simply put, programs had to know what generation of a specific file was being processed. This was called Generation Data Group (GDG) technology. Equitable had made three failed attempts at implementing GDGs, even though we were being led by our best and brightest.”

My father continued: “So in early 1973, while your mom, brother, sister and I lived in Easton, Pennsylvania, I was in New York for a monthly staff meeting, and our senior vice president asked to see me. He asked me about my GDG thoughts and I answered candidly. He then asked me to come back to New York and lead a new GDG attempt, a project likely to take several months. About four months in, I found myself in one of the high conference rooms of our Manhattan office overlooking Radio City Music Hall. It was way past dinner time and I had my 25-person project team around a huge table discussing how we were stuck in exactly the same spot as the previous three teams had been. Despite our best efforts and those of IBM’s best DB2 database technicians, we could not get the GDGs to conform to DB2’s elaborate requirements.”

Failure
He continued: “The team was mostly the same folks who were on the previous teams, and they viewed that failure as some kind of a self-fulfilling prophesy. I was trying to be an optimistic leader, but we’d been stuck there for weeks and on the inside I felt like a toilet bowl flushing.

“So I bailed out of the meeting and went to the bathroom to clear my head. I was standing at the urinal lost in thought, when my colleague Ed came in and stood at the urinal to my right. Ed was not what today’s crowd would call ‘the popular kid.’ I didn’t know him well, but I knew he was distant, quirky—brilliant—and I liked him. I always treated him with respect and he always greeted me with humor. On that night his greeting was: ‘What’s a big shot like you doing here so late?’

Ed was a database expert. I described our GDG roadblock to him. He went into deep thought, and then turned to me and said: ‘I think I can fix that.’

I returned to the conference room with him. The teams worked into the early hours, but by the following morning, we had a breakthrough.

Why hadn’t Ed been on the other GDG projects in the first place? Heck, why wasn’t he on my team?  Believe me, I asked. The New York leaders, knowing their personnel and their assignments, had made their ‘best’ available to the GDG projects. It seemed that Ed wasn’t viewed as one of the best, because he was different. There’s a lesson there somewhere, for all of us.”

Composition
The people equation often gets lost in IT. The lesson my dad taught me and the one that Kotch has learned while at AIG is this: Be open-minded, listen, and don’t be afraid to change the composition of teams—great ideas can come from anywhere.