Flexible Security in e-Manufacturing

At times, the industry falls back on just the "functional" description of e-manufacturing, viewing it as a software architecture that enables proactive improvement of overall factory effectiveness. This, indeed, is a major part of what it is, but this definition does not capture the grander concept that equates e-manufacturing with the continued and ultimate success of the semiconductor industry.

The semiconductor industry is challenged as never before with the ability to continue on the path defined by Moore's Law; the underlying needs have been clearly defined by industry leaders:

• Michael Splinter, president and CEO of Applied Materials, has said, "Staying on the path of Moore's Law by simple scaling will no longer work. Technology differentiation is being diminished. Neither semiconductor makers nor semiconductor equipment suppliers can individually afford the R&D load for technology development."
• Robert Doering, senior fellow at Texas Instruments, has expressed similar concerns, having observed that the International Technology Roadmap for Semiconductors (ITRS) indicates that many challenges the industry is facing in several technology areas are expected to hit "red brick walls" by the end of the decade. These walls represent discontinuities and will require novel and innovative solutions involving new materials and new processes.
• John E. Kelly, senior vice president and group executive director at IBM , states that the industry's challenge is to continue to drive CMOS technology as far as it can go, while in parallel developing new technology — nanoelectronics — necessary to keep on Moore's Law.

Alliances and partnerships are necessary to meet these challenges. Doering believes there is a greater need today for all sorts of partnerships for two reasons: the growing difficulty of the technical challenges that we face and the growing complexity and cost of emerging solutions. He also has expressed that alliances span a broad spectrum of activities; at one end are the long-range pre-competitive alliances, such as those under the umbrella of The Semiconductor Research Corp., and at the other end, more short-term, business-to-business partnerships intended to create focus, consensus and share costs for specific technology solutions.

Similarly, Kelly's view is that, as the industry moves to the regime of truly revolutionary technologies, partnerships will be crucial to generating ideas needed to address these challenges — and these partnerships may cross the boundaries of academia, industry (i.e., IC manufacturers, equipment suppliers and materials suppliers) and even nations.

Current examples of broadly based partnerships are nanotech clusters in Albany, N.Y.; Grenoble, France; and Quebec, Canada. Mehdi Moussavi, head of supplier relations at CEA-LETI in Grenoble, has stated, "To have a successful cluster initiative, you must bring together a very broad and diverse set of players into an alliance, including competitors." Regarding competitors, Splinter notes that what is pre-competitive to customers is competitive to one's peer group.

Another example of an emerging alliance paradigm focuses on design for manufacturing. In Kelly's view, this is the situation, as mask complexity is growing exponentially, requiring very fast, closed-loop processes to design, qualify and characterize masks — both initially and in reaction to design changes. Aart de Geus, chairman and CEO of Synopsys , has said that the days of independent semiconductor design are over, that design and manufacturing must be connected in a joint food chain that reflects "tech-onomies." Synopsys' announced joint-development project between design software and the production of advanced photomasks for 65 nm and below processes reflects this view.

So where does e-manufacturing fit? Electronic collaboration between partners is implicit in the operation of the kinds of partnerships discussed above. Moreover, implicit in electronic collaboration is the real-time exchange of data. A requirement, therefore, of any partnership or alliance is that data defined to be intellectual property (IP) by one constituent in the partnership must be secured from unauthorized use by other constituents. This protection must be in place to protect pre-existing IP being brought to partnership by its constituents, but also to bring protection and security to data evolving during the course of the relationship.

It goes without saying that IP is at the core in every enterprise in our industry. There is IP in process, design and manufacturing. We all have it; we all want to keep it. For some, licensing IP is the single revenue stream for the enterprise; consider Transmeta's 2005 announcement that it may move to a pure IP licensing business model. For others, the importance of IP has to do with patent protection (e.g., Toshiba's memory patent suit against Hynix). Time-to-market is also a core value proposition to IP protection (e.g., TSMC's and IBM's immersion lithography race).

E-manufacturing is the software architecture that enables the implicit electronic collaboration and real-time exchange of data. The catch is that partners have been reluctant to adopt e-manufacturing because, until now, security has not been associated with data content. Fortunately, a solution — Flexible Security Wrapper — has come out of a three-year NIST Advanced Technology project between ILS Technology, Advanced Micro Devices and Oceana Sensor Technologies . Briefly described, it is a distributed, object-based, openly architected software that associates security with data content rather than data transmission or data storage.