I remember when the first edition came out---back in the early 90's!!!! I was young, slim, had a full head of hair---and this new textbook was the bomb. Vastly better than any previous textbook on computer architecture.
I can't believe it is still the best. it's been like 30 years. During that time, so much has happened--the death of supercomputer companies like Convex and Cray, SIMD going from expensive computers like the MASPAR MP-1 to being on virtually every processor, the dot com boom, the rise of google-style server farms, etc etc.
And now the transition to neural net processing.
I mean, it is a testament to the authors that they could keep their competitors from even thinking about trying to write a competing book for so long. It is a great case study in how to stay relevant in tech for the long term.
But man, before it came out, every year 2 or 3 new textbooks in computer architecture came out, each one detailing the next cool thing which computer architectures were being called upon to do.
It's exhibit A of Peter Thiel's case that we are living in an era of very low innovation. If Computer Architecture were a really healthy field, classes would have to be taught from recently-published papers, because it was moving faster than a textbook could be published.
Hat's off to the authors, but man, this is really depressing.
> If Computer Architecture were a really healthy field, classes would have to be taught from recently-published papers, because it was moving faster than a textbook could be published.
I really don't get this perspective. How can you possibly hope to understand "recently-published papers" without first understanding the basics of the field, which is what Hennessy and Patterson covers? Every subject has introductory textbooks from which introductory courses are taught, and then you can take more advanced courses that can, among other things, include material from recently-published papers. Are there even any fields where courses must be taught from recently-published papers?
On another note, it's not like no more computer architecture textbooks are made. Look at the Synthesis Lectures on Computer Architecture series.
// I really don't get this perspective. How can you possibly hope to understand "recently-published papers" without first understanding the basics of the field, //
A healthy field moves faster than the textbooks can keep up. Consider the field of, say, personal transportation, during the turn of the century. I'm sure they had some very classic textbooks on veterinary medicine for horses and oxen. But they probably didn't have a lot of classic textbooks on being an auto mechanic.
In the 80's and 90's, if you were studying AI in school, they'd be teaching you prolog, first order logic, various parsing techniques, heuristics to search through graphs. Now it is neural nets, transformers, etc.
A healthy field has to "tear up the textbooks" every 10 years or so.
I can't believe it is still the best. it's been like 30 years. During that time, so much has happened--the death of supercomputer companies like Convex and Cray, SIMD going from expensive computers like the MASPAR MP-1 to being on virtually every processor, the dot com boom, the rise of google-style server farms, etc etc.
And now the transition to neural net processing.
I mean, it is a testament to the authors that they could keep their competitors from even thinking about trying to write a competing book for so long. It is a great case study in how to stay relevant in tech for the long term.
But man, before it came out, every year 2 or 3 new textbooks in computer architecture came out, each one detailing the next cool thing which computer architectures were being called upon to do.
It's exhibit A of Peter Thiel's case that we are living in an era of very low innovation. If Computer Architecture were a really healthy field, classes would have to be taught from recently-published papers, because it was moving faster than a textbook could be published.
Hat's off to the authors, but man, this is really depressing.