?? 00000010.htm
字號:
sive). So, although the initial design is often not in the critical path to <br />product deployment, redesign of the computer system may need to be done quic <br />kly to resolve problems. <br />Design challenge: <br />· Rapid redesign to accommodate changing form factors, control algorithms, <br />and functionality requirements. <br />6.3. Product families <br />In many cases embedded system designs are not unique, and there are a variet <br />y of systems of various prices and capabilities forming a product family. To <br /> the extent that system designers can reuse components, they lower the total <br /> cost of all systems in the product family. <br />However, there is a dynamic tension between overly general solutions that sa <br />tisfy a large number of niche requirements, and specifically optimized desig <br />ns for each point in a product family space. Also, there may be cases in whi <br />ch contradictory requirements between similar systems prevent the use of a s <br />ingle subsystem design. In the Mission Critical and Small examples different <br /> customers require different interfaces between the embedded system and thei <br />r equipment. In the Distributed example regulatory agencies impose different <br /> safety-critical behavior requirements depending on the geographic area in w <br />hich the system is deployed. <br />Design challenge: <br />· Customize designs while minimizing component variant proliferation. <br />7. Design culture <br />Design is a social activity as well as a technical activity. The design of d <br />esktop computers, and CPUs in particular, has matured in terms of becoming m <br />ore quantitative in recent years. With this new maturity has come an emphasi <br />s on simulation and CAD tools to provide engineering tradeoffs based on accu <br />rate performance and cost predictions. <br />Computer designers venturing into the embedded arena must realize that their <br /> culture (and the underlying tool infrastructure) are unlike what is commonl <br />y practiced in some other engineering disciplines. But, because embedded sys <br />tem design requires a confluence of engineering skills, successful computer <br />designers and design methodologies must find a harmonious compromise with th <br />e techniques and methodologies of other disciplines as well as company manag <br />ement. Also, in many cases the engineers building embedded computer systems <br />are not actually trained in computer engineering (or, perhaps not even elect <br />rical engineering), and so are not attuned to the culture and methodologies <br />of desktop computer design. <br />7.1. Computer culture vs. other cultures <br />A specific problem is that computer design tools have progressed to the poin <br />t that many believe it is more cost-effective to do extensive simulation tha <br />n build successive prototypes. However, in the mechanical arena much existin <br />g practice strongly favors prototyping with less exhaustive up-front analysi <br />s. Thus, it may be difficult to convince project managers (who may be applic <br />ation area specialists rather than computer specialists) to spend limited ca <br />pital budgets on CAD tools and defer the gratification of early prototype de <br />velopment in favor of simulation. <br />Design challenge: <br />· Make simulation-based computer design accessible to non-specialists. <br />7.2. Accounting for cost of engineering design <br />One area of common concern is the effectiveness of using engineers in any de <br />sign discipline. But, some computer design CAD tools are very expensive, and <br /> in general organizations have difficulty trading off capital and tool costs <br /> against engineering time. This means that computer designers may be deprive <br />d of CAD tools that would reduce the total cost of designing a system. <br />Also, in high-volume applications engineering costs can be relatively small <br />when compared to production costs. Often, the number of engineers is fixed, <br />and book-kept as a constant expense that is decoupled from the profitability <br /> of any particular system design, as is the case in all four example systems <br />. This can be referred to as the "Engineers Are Free" syndrome. But, while t <br />he cost of engineering time may have a small impact on product costs, the un <br />availability of enough engineers to do work on all the products being design <br />ed can have a significant opportunity cost (which is, in general, unmeasured <br />). <br />Design challenge: <br />· Improved productivity via using tools and methodologies may be better rec <br />eived by managers if it is perceived to increase the number of products that <br /> can be designed, rather than merely the efficiency of engineers on any give <br />n product design effort. This is a subtle but, in practice, important distin <br />ction. <br />7.3. Inertia <br />In general, the cost of change in an organization is high both in terms of m <br />oney and organizational disruption. The computer industry can be thought of <br />as being forced to change by inexorable exponential growth in hardware capab <br />ilities. However, the impact of this growth seems to have been delayed in em <br />bedded system development. In part this is because of the long time that ela <br />pses between new technology introduction and wide-scale use in inexpensive s <br />ystems. Thus, it may simply be that complex designs will force updated CAD t <br />ools and design methodologies to be adopted for embedded systems in the near <br /> future. <br />On the other hand, the latest computer design technologies may not have been <br /> adopted by many embedded system makers because they aren't necessary. Tool <br />development that concentrates on the ability to handle millions of transisto <br />rs may simply not be relevant to designers of systems using 4- and 8-bit mic <br />roprocessors that constitute the bulk of the embedded CPU market. And, even <br />if they are useful, the need for them may not be compelling enough to justif <br />y the pain and up-front expense of change so long as older techniques work. <br />That is not to say that new tools aren't needed, but rather that the force o <br />f cultural inertia will only permit adoption of low-cost tools with signific <br />ant advantages to the problem at hand. <br />Design challenge: <br />· Find/create design tools and methodologies that provide unique, compellin <br />g advantages for embedded design. <br />8. Conclusions <br />Many embedded systems have requirements that differ significantly both in de <br />tails and in scope from desktop computers. In particular, the demands of the <br /> specific application and the interface with external equipment may dominate <br /> the system design. Also, long life-cycles and in some cases extreme cost se <br />nsitivity require more attention to optimization based on these goals rather <br /> than maximizing the computational throughput. <br />The business and cultural climates in many embedded system design situations <br /> are such that traditional simulation-based computer design techniques may n <br />ot be viable in their current form. Such methodologies may not be cost-effec <br />tive given constraints on categories of expenditures, may not be seen as wor <br />thwhile by non-computer-trained professionals, or may simply be solving the <br />wrong problems. <br />Recent interest in hardware/software codesign is a step in the right directi <br />on, as it permits tradeoffs between hardware and software that are critical <br />for more cost-effective embedded systems. However, to be successful future t <br />ools may well need to increase scope even further to include life-cycle issu <br />es and business issues. <br />The tutorial slide presentation presented at the conference augments this pa <br />per, and may be found at: <a href="http://www.cs.cmu.edu/~koopman/iccd96">http://www.cs.cmu.edu/~koopman/iccd96</a> <br /> <br />-- <br />※ 來源:·BBS 水木清華站 smth.org·[FROM: 202.204.8.215] <br /><a href="00000009.htm">上一篇</a><a href="javascript:history.go(-1)">返回上一頁</a><a href="index.htm">回到目錄</a><a href="#top">回到頁首</a><a href="00000011.htm">下一篇</a></h1></center><center><h1>BBS 水木清華站∶精華區</h1></center></body></html>
?? 快捷鍵說明
復制代碼
Ctrl + C
搜索代碼
Ctrl + F
全屏模式
F11
切換主題
Ctrl + Shift + D
顯示快捷鍵
?
增大字號
Ctrl + =
減小字號
Ctrl + -