-------------------------------------------------------------------------------------------- NOTE: THIS IS A SAMPLE REVIEW PROVIDED BY A STUDENT. It is not an instructor's solution, but the instructor likes this review due to its overall clarity and thoroughness. Certain parts of the review, such as grammar, can certainly be improved, and the review can be made even more thorough and clear, but we are providing you this as is so that you have a good example solution in your hands that shows our expectation. -------------------------------------------------------------------------------------------- Reviewed paper: "Requirements, Bottlenecks, and Good Fortune: Agents for Microprocessor Evolution" by Patt Student reviewer: Name withheld to protect the diligent. == Paper summary == The development of microprocessors from the 70's was due to multiple contributing agents within the context of computer architecture tradeoffs. The contributing agents for evolution are 1) new requirements; 2) bottlenecks; and 3) good fortune. These agents have affected how tradeoffs are made by engineers and researchers within the basic framework such as levels of transformation, or design points. After understanding these agents and tradeoffs, one can predict what's to come in the area of microprocessor in the future. == Strengths of paper and mechanisms == * The framework summarized in this paper becomes important groundwork for the analysis of the agents and the predictions of the future developments of microprocessors. * This paper also summarizes some of the recent technologies in the area of computer architecture, presented in a way that is easy for people having little background to understand. * The paper also extracted the three ingredients or agents for future evolutions of microprocessors. * The predictions in the paper are mostly correct and accurate, where some of them are still yet to be realized, providing useful predictions for the future. == Weaknesses of paper and mechanisms == * The paper presents the three agents for evolution without explaining in depth how those agents as a whole pushed the developments of microprocessors. * The 'Good Fortune' of the three agents needs some more explaining about why it is one of the three agents. == Detailed comments == The development of microprocessors has advanced at an amazing speed for a few decades. This paper is dedicated to discussing why such advancement is possible, abstracting the reasons behind the scene and predicting the future in this area with those reasons. Written by a computer architect, the analysis is designed to be systematic and logical. The author first provides a basic framework and lay out some of the important tradeoffs in computer architecture. From that, he proposed the three agents for advancements and argues that those agents are well within the context of the above-mentioned framework. Then the author summarizes the developments of microprocessors from the 70's to the beginning of the 21st century, suggesting that the three agents are the driving factors for these new technologies. After validating the thesis of the paper with a few examples, the author goes on predicting the future developments of microprocessors and how computer architects will make more interesting tradeoffs based on the three agents mentioned before. == Ideas for improvement == Relate back to the three agents more often from both the summary and the prediction sections. This will make the thesis more apparent to the readers. Explain better why good fortune is one of the agents, with real examples about how past research fails without good fortune. In the prediction section, provide some guidelines on what people should do with the changes to come. == Lessons learned == By reading this chapter, I got a better understanding of the history of microprocessor and a sense of 'how we got here'. After reading the three agents of evolution, I understood some of the reasons behind this incredibly fast advancement in the microprocessor industry. While most of the predictions such as System-on-Chip, and multilevel caches have come into reality, others have yet to be widely used. Thus by reading this paper, I realized that there are still a lot of opportunities in this area, waiting to be discovered and improved. -------------------------------------------------------------------------------------------- NOTE: THIS IS A SAMPLE REVIEW PROVIDED BY A STUDENT. It is not an instructor's solution, but the instructor likes this review due to its overall clarity and thoroughness. Certain parts of the review, such as grammar, can certainly be improved, and the review can be made even more thorough and clear, but we are providing you this as is so that you have a good example solution in your hands that shows our expectation. -------------------------------------------------------------------------------------------- Reviewed paper: "Requirements, Bottlenecks, and Good Fortune: Agents for Microprocessor Evolution" by Patt Student reviewer: Name withheld to protect the diligent. == Summary == This paper's stated goal is threefold: first, to provide a basic overview of the microprocessor design process with a particular emphasis on architecture, second, to discuss historical developments of microprocessors over the thirty years spanning 1971-2001, and third, to speculate on possible developments in the "coming" years c. 2001. The paper's discussion is centered on three main "agents for microprocessor evolution" identified by the author: 1) new requirements or desired user applications, 2) bottlenecks in existing systems that can be alleviated by architectural modifications, and 3) good fortune, including, for instance, microelectronic innovations and process node shrinks. To me, these are the most relevant aspects of the paper today as they are relatively timeless. When the paper was written, the speculation on impending architectural developments may have been more significant; in retrospect, they chiefly serve to underscore the author's predictive accuracy. == Strengths == * Good high-level, easily-readable historical background on processor development – seeing where we've come from * Timeless "agents for evolution" – thinking about where we're going * Predictive accuracy – notable examples include IV(A)'s talk of increased use of reconfigurable logic -> today's FPGAs and products such as Xeon Phi, IV(E)'s thoughts on intelligent clock use (a la Intel's Turbo Boost), and so forth == Weaknesses == * Somewhat dated content – despite the accuracy of the predictions, their utility is limited over thirteen years post-publication == Personal impact == I was particularly struck by the accuracy of the author's predictions on architectural developments in microprocessors over the "next" decade (the '00s). At this point, I don't have nearly the body of knowledge required to make such predictions myself about likely future innovations. However, I'd be interested to look back at this paper at the end of this semester and make some speculations myself based on his three "agents for microprocessor evolution" and the knowledge I will gain throughout this course. -------------------------------------------------------------------------------------------- NOTE: THIS IS A SAMPLE REVIEW PROVIDED BY A STUDENT. It is not an instructor's solution, but the instructor likes this review due to its overall clarity and thoroughness. Certain parts of the review, such as grammar, can certainly be improved, and the review can be made even more thorough and clear, but we are providing you this as is so that you have a good example solution in your hands that shows our expectation. -------------------------------------------------------------------------------------------- Reviewed paper: "Requirements, Bottlenecks, and Good Fortune: Agents for Microprocessor Evolution" by Patt Student reviewer: Name withheld to protect the diligent. Paper Summary: The paper outlines the agents that have driven the evolution of microprocessors and how that evolution has actually panned out. Patt claims that the first of three driving forces of microprocessor change is new requirements. Such requirements have included the need for higher performance, multiple instructions being processed at once, and power consumption/awareness. The second factor is bottlenecks in the data supply, instruction supply, or instruction execution. Patt explains that the bottleneck of slow memory was overcome by caching, and the bottleneck of conditional branching was over- come by out-of-order instruction and branch prediction. The final agent of evolution is simply good fortune. Patt also highlights how the evolution has actually occurred from 1971 to the present. Moreover, he discusses the future of microprocessors and how it will likely include algorithms embedded in the processor that solve problems in the application space i.e. more application-aware microprocessors. He also predicts that new materials, more internal fault tolerance and a number of other developments will solve problems of current microprocessors. Strengths: • The paper is well structured and appropriately addresses the topic of past microprocessor evolution. • The paper is a good, concise summary for anyone who wants to under- stand how the microprocessor has evolved in the last few decades. • As a computer architect should, the author looks forward (future technologies), backward (past developments), up (at the application space), and down (at the materials and circuit levels). Weaknesses: • As it was supposed to be the focus of the paper, the discussion of the actual agents of evolution was surprisingly short. This was especially the case for good fortune section, which was not adequately explained. • Rather than simply stating the possible future developments, the author could have discussed how the agents of evolution that he discussed might drive each of new developments. • The article would be difficult to understand for someone without a good working knowledge of computer systems. Ideas for improvement: • The scope of the target audience could be broadened by briefly defining some concepts mentioned in the paper. For example, clock skew could be quickly defined in parentheses, making the discussion of clock skew clearer and more accessible. • Expand on the explanation of good fortune as an agent of evolution. Lessons Learned: The paper gave me a good overview of the timeline of microprocessor development. I have a better understanding of the challenges faced by computer architects, so now I can make more informed decisions regarding design trade- offs. I also now know some of the bottlenecks I should be aware of when designing my own architecture. Furthermore, I have learned that there is much more to be done in the realm of microprocessors and there will be a lot of opportunity to make in advances in the field in the future. Finally, if I work in computer architecture in the future, I now have a better idea of what technologies will likely make these advancements.