计算机专业留学个人陈述.pdf

计算机专业留学个人陈述 Ｏf all the areas of computer science ， parallel processing fascinates me the most. Ｔhe term first came to me when Ｉ read the book Ｃomputer Ｏrganization Ｄesign： the Ｈardware/ Ｓoftware Ｉnterface as part of my Ｃomputer Ａrchitecture course. Ｂy teaching me how the performance of computers can be improved through pipelining and parallel processing， the book reshaped my thinking in computer science. Ｍy lab experience deepened my understanding of parallel processing. Ｉn my third year， Ｉ worked on a ＧＩＦＴ project with Ｄr. Ｋomatsu， chief researcher of ＩＢＭ Ｔ okyo Ｒesearch Ｌaboratory and Ｄr. Ｋoseki， one of Ｄr. Ｋ omatsuaposaposs colleagues. Ｔ he project was to introduce the architecture supports that can allow compilers to derive more parallelism from programs. Ｔ hese architecture supports include three designs， a conditional execution mechanism to execute instructions without normal conditional jump instructions ， speculative execution mechanism， and a dynamic memory disambiguation mechanism to execute memory access instructions simultaneously. Ａs undergraduate students seldom get to be involved in actual 1 / 3 research， Ｉ took full advantage of the opportunit of working in the lab. Ｉ particularl enjoyed the seminars held twice a week as part of the lab work.Ａt these seminars， Ｉ engaged in in-depth discussions with Ｍasteraposaposs students under Ｄr. Ｋomatsuaposaposs seasoned guidance. Ｔhe lab experience is adding weight and insights to m graduation paper Ａn Ｉmprovement of Ｓoftware pipelining using Ｌoop Ｔransformations. Ｉn this paper ， Ｉ will introduce a translation algorithm using loop interchange and loop