These couple of weeks I’ve been on the road again, speaking at TechEds as well at universities across the Asia-Pacific region. Last week I was at TechEd South-East Asia and also spoke at a faculty summit in Kuala Lumpur as well as at the National University of Singapore (NUS). This week, I’m in Taipei for TechEd Taiwan and related activities. At most events, people have been somewhat surprised (most of them pleasantly) to know that a core mission-critical product such as SQL Server is setting up an R&D Center in China and are curious to know the background of why we decided to build the team and what kind of work we’ll be doing there. I thought it would be a good idea to explain some of this here.

First, some history J. It was in the early 90’s that Microsoft got serious about the database business and started assembling a top-notch team in its SQL Server division. There were names like Hal Berenson, Peter Spiro and Dave Campbell (my current boss) who joined from DEC, Bill Baker from IRI, James Hamilton from IBM, Anil Nori from Oracle, Pedro Celis from Tandem and several other leading minds from virtually every corner of the database industry. And of course on the research side, there were names like Jim Gray, Dave Lomet, and Phil Bernstein among others. It was this set of people that took ownership of SQL Server and transformed it from being a bit-player in the database market to the leadership position that it now occupies today.

Throughout this time, until about 1995, all SQL Server development was done entirely in Redmond, Washington. We did this because, essentially, we could. Microsoft could hire the talent it needed – whether from academia or industry, mostly from the US but also from other locations – and bring them to Redmond. And if you can do this, there are certainly some advantages to having everyone co-located in one place: if you have a problem with a piece of code, you walk down the hall to talk to the person who wrote it; if you need to call a meeting, you grab the people you need and walk over to the coffee shop to hash out the matter. So it was a great model, while it lasted.

In the last few years, however, a couple of important things began to change. First was the geographical distribution of where talent resided, now and especially in the future. It has been true for some years now that countries outside the US, especially in Asia, have produced a lot more engineers than the US itself, and that gap has been widening every year. Increasingly however, rising standards of living in Asia coupled with increased difficulty in obtaining H1-B visas and Green Cards for foreign nationals in the US have resulted in many more of these engineers staying in their home countries instead of moving to the US as had previously been the norm. It was clear that in the future we were increasingly looking at a world where world-class talent was more dispersed around the world, especially in countries such as China and India.

The second factor that changed was the increasingly geographically diverse nature of our partner and customer-base. As SQL Server became more and more successful, we started seeing increasing numbers of partners and customers, including some very large ones, in regions outside the US. It is hard for a development team located in a single place, no matter how smart, to be able to understand the needs and desires of this global partner customer-base. Thus, it was no longer clear that it was the best policy to locate the entire team in one place, even if we could do so.

It is in this context that the SQL Server leadership team looked at the situation in 2005 and made the decision to expand our R&D footprint beyond Redmond by creating two additional strategic R&D sites, one in India and another in China. While this might seem like an easy decision to make, the hard part is actually in the execution. How do you take a complex product such as SQL Server and build R&D teams for it in these new locations? Yes, it is true there is a lot of talent in India and China, but it is also true that a lot of it is what I call “raw talent” – very bright, but lacking deep experience in developing commercial-grade products, especially systems-level platform products. It takes a lot of effort and energy and experiences to transform an individual from being a “smart coder” to being a “professional engineer” with all the implications that term has for me – someone who is at the top of their game with respect to translating ideas into mission-critical industrial-strength code in the most efficient and effective manner possible. It also takes a lot of time, effort, energy, experiences and most importantly, commitment to transform a set of very talented individuals into a world-class engineering team. It is not an easy challenge; but then again, if it were easy it would not be much fun, would it?

But that is precisely what we have set out to do – build a world-class engineering team that works on some of the broad and exciting challenges in the Data Platform world that I described in my previous post. So we investing in several areas in the China R&D team – in our core Data Access technologies, in XML, in BI, in Tools, and in some very new and exciting areas that we are not ready to share externally just yet. I’ll discuss more details about some of these areas in future posts, but clearly we’ll be creating products and technologies for the world market out of our China team.

But we are going beyond that. Part of the reason for being in China, as I mentioned before, is to be able to connect more deeply with our increasing base of customers and partners in this region. So the China team will also focus on creating products and technologies that address the specific needs of the local market, in China specifically, but more generally for the entire Asia-Pacific region. In the long run, this is actually very exciting both for the team and for our business in this region.

Make no mistake, this is going to be a long and sometimes difficult journey. There are many challenges that the team has to tackle and overcome - technical challenges, process challenges, communication challenges and so on. But perhaps the most important and interesting leadership challenge is that of culture – how do we make sure that we build a team that has the “right” culture? And what exactly is that culture? Is it the same as the culture the team in Redmond has? Or does it have to be customized and adapted in some ways to the location and context of this team in China? These are very interesting questions and ones I believe are going to fundamental to the success or otherwise of our team and others like it. I’ll talk about some of these issues in a future post. For now, let me end by saying that we are very clear in this goal – we will build the Premier Data Platform R&D team in the Asia-Pacific region, and we are really excited to be on this journey.

Until next time – cheers and zai-jian!

Prakash – 孙博凯

Posted Monday, September 17, 2007 3:22 PM by prakas
------------------------------------------------------------------------------------------------------------