PDF
Molecular Computing
Edited by Tanya Sienko, Andrew Adamatzky, Nicholas G. Rambidi, and Michael Conrad
MIT Press, 2003, $45.00, ISBN: 0-262-19487-2
The apparent inevitability of Moore’s law teaches us that the limit of VLSI (very large scale integration) in silicon chips is fast closing upon us. In a decade or so, we will be forced to look into new approaches in circuit design, whether rooted in quantum mechanics or molecular biology. Molecular Computing offers a survey of the techniques being explored in the latter.
The book’s seven loosely joined chapters address the major aspects of this field. The first chapter (written in part by the late Michael Conrad, to whom the book is dedicated) suggests that proteins and other biological molecules should be considered as building blocks in future information processing systems, thanks to their ability to recognize various molecular patterns.
Chapter 2 further addresses the recognition capabilities of various macromolecules. The next chapter introduces the reaction-diffusion process, in which data is encoded as concentration profiles of specific reagents, the diffusion waves of which can then be seen as performing computation.
Chapter 4 addresses the complexity issues related to this new computing paradigm, together with additional algorithms. One of the stars of molecular computing is DNA (deoxyribonucleic acid) computing, pioneered by Leonard Adleman’s use of DNA’s self-annealing ability to find a solution to the Hamiltonian path problem. Addressing such problems has led to frantic new research activities, surveyed in chapter 5.
The last two chapters focus on bioelectronics, or the “wetware” technologies required to implement the designs introduced in the previous chapters.
I had hoped this would be the definitive textbook on this topic (although “definitive” is probably too strong a word for a discipline still in its infancy), but this is not the case. The book lacks continuity and homo-geneity. Some aspects are either repeated or introduced too late and the prerequisites for the different chapters vary wildly.
However, to someone with a good knowledge of the basics of chemistry, molecular biology, and electronics, this book provides details on the latest developments in an exciting new field that shows promise for the future of the computing industry.—Shrisha Rao
Embedded Software for SoC
Edited by Ahmed Amine Jerraya, Sungjoo Yoo, Norbert When, and Diederik Verkest, Kluwer Academic Publishers, 2003, $174.00, ISBN: 1-4020-7528-6
This volume is an edited collection of papers from the Embedded Software Forum of the DATE (Design, Automation, and Test in Europe) conference held in March 2003. The 37 papers focus predominantly on practical systems issues. The organization of the topics into 10 sections corresponds to the sessions of the conference, but this structure proves ungainly.
Several papers look at the role of abstraction layers and interfaces in embedded software. Topics include generating embedded software from a specification; investigating mapping and automatic code generation of applications to a hardware platform and its realtime operating system; and experiences with formal methods applied to automotive software applications.
One section looks at modeling hardware and software components of embedded systems. Papers investigate software performance using transaction-level modeling, as well as the scheduling and timing of hardware/software communication for on-chip multiprocessors.
The volume includes papers exploring source-code-level control flow optimization; software streaming techniques; and system-level verification using simulation.
Because embedded systems often must conserve power, energy-aware design may prove critical to maintaining battery life. Several papers address this important issue.
This timely collection provides readers with a breadth of recent contributions in this rapidly changing field. Although the speed in completing this volume within months of the DATE conference introduced some production problems, this collection of papers will serve as a useful reference to those interested in software for embedded-systems design.—Gregory D. Peterson
Reprinted from Computing Reviews, © 2004 ACM, http://www.reviews.com
Originally published in Queue vol. 2, no. 4—
Comment on this article in the ACM Digital Library
© ACM, Inc. All Rights Reserved.