Chris Christopher joined HP in 1968 as a research and development engineer in the former Calculator Products Division. In 1975, he managed development of desktop computer software, and two years later became R&D manager for the desktop computing division. After several other promotions, he became general manager of the Workstation Systems Division in 1997. A native of Greece, he received bachelors and masters degrees in electrical engineering at Colorado State University.
For engineers to be effective, and for them to efficiently meet their time-to-market requirements, they must work within a value chain that's seamless in terms of computer platforms, so their software applications can run smoothly, says Christopher.
DESIGN NEWS: What decisions must engineers make when thinking about their computer needs?
Christopher: It's vitally important that engineers and the management team understand what architecture they should have for the work they do. They have to decide where they want to go, how the work flow will be, the capabilities they want. Those decisions will help them decide what hardware and applications they should have in their work.
Q: Should they think about the value chain they work within?
A: Yes. The value chain is chopped up and segregated. To achieve faster time to market, greater effectiveness, and to satisfy customers, the value chain will have to be seamless. You can't just buy a toolset that's state of art and expect to shorten time to market. Once the engineer designs the part, what do you do with the database? He has to go through the supply chain. Will others in the supply chain have a different database? The customer will want replacement parts. Will he have a different database? The database the engineer creates can't stop at his desk. The implications for computer hardware are that vendors need a line of products that address different needs, like workstations and servers. If you buy a workstation from one vendor and a server from another, it isn't seamless.
Q: Should engineers buy their applications first or their hardware first?
A: They should look at the total solution. They should first ask what they think they need. If you buy hardware, then an application, then a database, you break the value chain. Focusing on one tool at a time is easy, but not always the best choice. First, think through your needs.
Q: What's the future of the Linux operating system in engineering?
A: Linux will be a building block. Already some CAE applications in mechanical and electrical design are on Linux. The government is interested in Linux for creating big clusters to solve large problems. Clusters give you the ability to scale for your high performance applications cost-effectively.
Q: Does UNIX have a future in engineering?
A: Yes it does. A lot of companies buy UNIX systems today and will continue to do so. There are a lot of applications for UNIX, and the platform is robust. UNIX servers will continue to grow over the next three to five years. UNIX is good for a lot of big analytical problems. Also, there are a lot of legacy applications in companies that run on UNIX. Enterprise customers have mission-critical environments on UNIX, and they don't take change lightly.
Q: What's the next breakthrough in computer architecture?
A: It's the IA-64 architecture. It represents the first time the engineer will be able to run multiple environments on one box. Engineers use workstations only 20% of the time. They use Windows the rest of the time. A multiple environment helps the engineer. With the IA-64 architecture, computer power will be a utility.