文章吧手机版
《Modern Operating Systems》读后感摘抄
日期:2022-05-14 12:12:41 来源:文章吧 阅读:

《Modern Operating Systems》读后感摘抄

  《Modern Operating Systems》是一本由Andrew S. Tanenbaum著作,Pearson出版的Hardcover图书,本书定价:USD 132.00,页数:1104,特精心从网络上整理的一些读者的读后感,希望对大家能有帮助。

  《Modern Operating Systems》精选点评:

  ●少有的读起来不让人想睡觉的《操作系统》教材

  ●第一次讀完,還有許多不懂之處,以後有機會再研究。

  ●跪拜Jason

  ●好多私货 仍需再战

  ●书本很简明很容易懂~最主要的是因为是我的课本!!!

  ●很喜欢这本书,上个学期读完的,I/O 和 Filesystem 还差一点点,Tanenbaum 总是能够用风趣的叙事把看起来「复杂」的东西讲清楚。所以这本书的相同内容虽然比耶鲁那本恐龙书薄很多,却料并不少。Tanenbaum 总是强调过时的算法总是会以意想不到的方式「复生」;仔细想想,这便是我们要研读历史的原因吧。

  ●1000多页的废话啊。。。作者的英语有待修炼。

  ●囊括linux, vista和symbian

  ●完全不晦涩的一本教材 很容易看进去

  ●小本影印版特别亲切

  《Modern Operating Systems》读后感(一):为什么没有勘误表?

  为什么没有勘误表?

  在亚马逊上看到外国人评论这本书的英文版,

  说里面有很多错误。我自己也看到一些错误。

  为什么没有勘误表?

  为什么没有勘误表?

  为什么没有勘误表?

  为什么没有勘误表?

  为什么没有勘误表?

  在亚马逊上看到外国人评论这本书的英文版,

  说里面有很多错误。我自己也看到一些错误。

  为什么没有勘误表?

  为什么没有勘误表?

  为什么没有勘误表?

  为什么没有勘误表?

  《Modern Operating Systems》读后感(二):理论性比较强,很好的操作系统教材

  配图很精髓,对理解帮助非常大。

  行文风格诙谐幽默,阅读时不易枯燥。

  在不同方法实现的后面,会给出作者经常会结合实际的应用给出评价,能让读者更深刻理解每种实现方法的利弊。

  不足之处是,全书篇幅较大,略考验读者的毅力。

  各别section的内容组织略显冗余。

  我买的是机械工业出版社的英文第三版,有极少量印刷错误(自己发现的,并不确定)

  (1) P.200, 第一行中PT2=2,感觉该是PT2=3

  (2)P.341,第三段的Fig1-6(a),应该是Fig1-7(a).

  综合来说,操作系统的经典教材

  《Modern Operating Systems》读后感(三):介绍操作系统的书籍中这本最高

  我念的是原版的英文,这本书不光介绍概念十分清晰和富有条理,而且作者用词引人入胜,用了许多口语化的单词,不会让人觉得枯燥和乏味,封面的设计也是非常有意思,没事看看封面想想这些概念,也挺有启发性。许多地方作者都用了生动形象和贴切的比喻,让人感觉思路十分明朗,不会被复杂的概念搞得一头雾水。状态好的时候,能够令人读的津津有味手不释卷。而且繁简适当,不会显得语句罗嗦,更不会觉得作者解释的不够和不清楚,因为他讲述一个步骤和算法的时候,必定会采用一个比较典型实际的例子加以重述,不会让人有“这边看不懂,下面就没有重复了怎么办?”的困扰和担忧。

  而且本书最富有魅力最值得赞赏的地方在于——作者引入一个概念必定是由一个问题和困境带出的,也就是说,作者不会无缘无故的列提纲似的推出一个全新概念让人摸不着头脑思索为什么,而是由“既然有这样的情况,那我们迫切需要达到的效果是什么,由此我们自然而然地想出了那样的实现和解决方式”这样的思维模式来引导读者,循序渐进的方法令读者不会一下子脱离进度。这能够让人很连贯的一气呵成的读下去,学会科学的思考方式。这一点与同类型的《Operating Systems Concepts》相比就好得多了,而且从图片上来看,这本书的图片也非常好懂,脱离了抽象的硬件形象,而采用几乎是脑海中的想象模式来演绎步骤。令人印象深刻也不显得乏味。

  这么好的书作者当然是很强大的,他出品了包括《Computer Networks》在内的许多书籍,在亚马逊的评价也是颇高,2008年2月更是紧跟时代的步伐推出了第三版3e,其中特地包含了对当时较新的Vista系统的解释。价格也是同类书籍中最高的。所以我购买了International Edition,就是所谓的山寨版,价格也要20镑左右。

  :那天作者Tanenbaum来我们学校做演讲,他瘦瘦的,有些秃顶,本人也是非常有趣的,介绍的话题是MINIX 3,其中提到MINIX 3的logo是一只浣熊,“为什么是浣熊呢?”——“因为他eat bugs""more likely to visit your house than a penguin" 下面大家都笑了……连我们的操作系统老师也要特意向他提问呢!

  《Modern Operating Systems》读后感(四):全面细致,深入浅出;兼心得、习题解答、勘误

  终于读完了,有种如释重负的感觉。

  原书篇幅不小,作为一个英语并非强项的读者来说,不时需要查查词典甚至对照下中文版来阅读的。

  其实书背上关于作者Tanenbaum的介绍,唯独缺了一条:MINIX最初的作者!这点是我在读时看到一句“MINIX是由本书的作者写的”这样轻描淡写的话时才意识到的。

  各种优点与其他书评重复的就不提了,我做完了前8章基本全部的习题(除去一些比较耗费时间的编程题),其中不乏大量极具启发性的。

  阅读时收获的确不小,特此把笔记整理如下,其中包含一些读书思考、比较好的习题的解答分析、以及个人完成的勘误(含部分中文版勘误),希望对其他的读者有所帮助:

  《Modern Operating Systems》读后感(五):Most Influential yet Misleading Textbook

  As someone has noted, I don't know why high rate is given to this book, not only here, but also on Amazon. As a student who spent years on kernel source codes, I can pick tons of errors from it. Anyway, if your lecture or course or whatever requires it, ignore what I said here.

  efore fault picking we could start discussing why a CS student should learn operating system. As my experience, the insight that may help students building the basic knowledge about infrastructure and understanding applications based on it without inspecting into trivial details is of great significance. From this perspective, proper introduction on APIs and explanation on rationale behind them are preferred. However this book is entangled with too many details involving kernel implementation and just skims over them without discussion in depth. In one sentence, this book fails in both API introdction and kernel explanation.

  The most severe problem throughout this book is that the author NEVER DISTINGUISHES PROCESSES AND THREADS. If we do not dive deeply into how flags are passed to clone system call(man 2 clone for details), one difference is important for programmers, namely memory space. Processes don't share memory spaces. Skipping all details in paging system which is merely mentioned in a section in this book, a pointer in a process cannot access the variable in other processes. This results in tremendous differences in application designs adapting multiprocess or multithread model and it's a hintsight for the reason why there should be interprocess communication: Because two processes want to share some data, instead of "the output of first process must be passed to the input of second process" in section 2.3. Thus all information provided in that section is extremely chaotic. It mixes thread level race condition(mmap would be more complicated) into process level and always fails to point out that for languages like C, C++ or Java mutex is the key to implement a critical section. And mutex should BY NO MEANS be a normal integer in any user program else than library because it makes strong assumption on compiler and hardware, even with volatile keyword as C/C++ language standard doesn't dictate the atomicity of binary integer operations and read/write of any variables. In addition, mutex provided by library yields nice implementation of sleep and wake mechanism so programmers don't need to worry about signals and when to wake waiting process up. Splitting everything academically only misleads readers into mess. In section 2.3.3, even different languages are listed for different busy waiting examples. This is really pointless because any C/C++ will be compiled into assembly language afterall and why are both C version and assembly version provided?!

  Apart from previous mixture, author also fails to distinguish swapping and memory management. When talking about MM, this book is pretty much like ridiculous imagination. The memory mapping in paging system provides readers nothing but one table pointing to another without any logic, so as the segmentation.

  I simply wrote those I've read. There are too many non-frivolous points got wrong in this book. I'd recommend young students reading more codes than reading such textbook.

评价:中立好评差评
【已有2位读者发表了评论】

┃ 《Modern Operating Systems》读后感摘抄的相关文章

┃ 每日推荐