Successful Chief Technology Officers (CTO) Strike a Balance Between Technology and People

05/15/2014 11:06 am ET | Updated Jul 15, 2014

I recently had the great privilege and opportunity to speak with a pioneer in high performance computing, big data and predictive analytics, and the evolving role of chief technology officers (CTO) with Dr. Eng Lim Goh, CTO at SGI. InfoWorld named Dr. Goh one of the World's 25 Most Influential CTOs. Dr. Goh has been with SGI for 25 years and has spent the past 10 years in the role of CTO where he oversees technical computing programs with the goal to develop the next generation computer architecture for the new many-core era. His current research interest is in the emerging need for in situ or integrated visualization of Big Data.

Dr. Eng Lim Goh, CTO SGI

In our talk with Dr. Goh, I learned a lot about scientific computing, data and analysis, but what struck me the most was his demonstration of the importance of striking a balance between the technical and people sides of business. Dr. Goh spends half of his time travelling to meet with the field sales teams, as well as customers, such as NASA and PayPal, who are doing profound things. He is proof that the success of today's CTO is not just predicated by how deep and extensive their technical acumen is, but by their ability to include elements of social responsibility, business acumen, mentoring, coaching and customer engagement.

Dr. Goh provides us with a glimpse into the future of high performance computing and Big Data in the enterprise and advises on the importance of nurturing employee and customer relationships:

Allow yourself to be led to a more profound goal - Using scientific data analysis and predictive analytics in the enterprise can range from search to discovery. Search is where you have a very certain and deliberate goal, but your haystack is huge and you are trying to find the needle - watch this brilliant SGI video. Discovery is where there is higher uncertainty because you are not sure what the needle looks like and you are trying to get the machine to tell you what you are looking for. On either side of the spectrum, or anywhere in between, the key is to get to serendipity. Serendipity happens when you start with a goal and the machine guides you to a more profound goal.

Dr. Goh shared an example of what happened to a key inventor at Raytheon, who when asked to design a radar, walked into his lab one day and because someone had forgotten to switch off the radar, the chocolate bar in his pocket melted. This unintentional, yet profound, discovery led him to go on to build the microwave oven. "When looking at problems, consider how profound the question or problem is. Doing an analysis to confirm a prediction, but then discovering that it is totally different than you originally thought is profound," says Dr. Goh.

Integrate data analysts throughout the organization - Because data analytics is becoming a more standard part of the business, the different lines of business will need their own data support. According to Dr. Goh, in the future, we will see the roles of data analysts, data engineers and data scientists being spread throughout the organization, or centralized depending on company size. One thing is for sure, decentralized or centralized, data analytics is here to stay.

Use high performance computers to get better insights - In his video on technical computing, Dr. Goh explains that there a two types of high performance computing - data collection and modeling/simulation. Data collection deals with building big instruments that generate data and take data in so that analysis and simulations can be run to produce output. With modeling/simulation, there is no real massive data input to the system, computer models are built to try to simulate the natural world to see how close their output matches the real world.

Gartner says that in 2020 over 30 billion connected devices will be in use. Can high performance computers keep up with this massive explosion of stuff connected to internet? According to Dr. Goh, if you want to ask questions of the entire set of data to get better insights, then there will be a demand for this kind of machine. The good news is that you don't need to be a computer scientist to use these machines.

Dr. Goh has the privilege of working with Stephen Hawking, arguably one of the smartest people on earth. SGI built Hawking one of the largest machines, which looks like one huge laptop. Hawking says, "The reason I use this machine is that it allows me to be a scientist, not a computer scientist." Hawking has taught Dr. Goh to make machines with scientist and analyst friendly platforms. "When you approach this supercomputer, you feel like you are approaching your own laptop, only big at the back end," says Dr. Goh.

Be conscientious about what you store - As the collection of data grows, with sensors and everything, businesses will need to ask themselves, "Is it really worth it?", "Can we get more insights for discovery?" and "Do we just stream it or do we store it?" As storing everything becomes untenable, businesses will need to be conscientious about what they store. Dr. Goh advises that if you are not storing, it's best to bring in massive amounts of compute power to run all different angles on that data because you won't have a chance to go back to it. Instead of storing data, you can store the learnings.

Make time for mentoring - As CTO, Dr. Goh spends a lot of time engaging with younger engineers. He says this mentorship is very important and is a key tenant of the more senior engineering roles in his company. "Young engineers come in bright, enthusiastic and with lots of ideas, and when mentored young, they launch quickly," says Dr. Goh. The goal for the CTO is to filter innovations. Dr. Goh says that trying to figure out what to pick is the big problem. Talk about a good problem to have!

Spend time with customers - Last year alone, Dr. Goh had more than 150 customer engagements. He says there is no replacement to meeting with customers live and shaking their hand. So how does this outside-facing CTO filter all the feedback? Dr. Goh says that making that eye-to-eye connection to see how a customer presents their feedback allows him to tell how important that feedback is. He takes this feedback back, distills it and puts it through an internal filtering process. According to Dr. Goh, a great innovator once said that sometimes the customer knows what they want, but many times they do not know what they need. "If Henry Ford had asked his customers what they wanted, they would have said a faster horse," says Goh.

A customer may be very clear on what they want based on their current workflow, but may not understand all the technology coming that will change their workflow in the future. It is important to go in and really learn the customer's workflows, and after a few sessions to make suggestions and bring the feedback back to engineering. Sometimes the solution is not building a new product, but tweaking the product to be suited to the new workflow.

You can watch the full interview with Dr. Eng Lim Goh here. Please join me and Michael Krigsman every Friday at 3PM EST as we host CXOTalk - connecting with thought leaders and innovative executives who are pushing the boundaries within their companies and their fields.