Systems Performance (Addison-Wesley Professional Computing Series)

Simply the most use performance engineering book that I own. I already had the first edition and had to buy this second edition. Its full of practical, real world advice for tuning modern Linux systems. It's readable, comprehensive and authoratitive.

I am a software engineer who does systems performance analysis as part of my job. This book let me grow my knowledge, it did expand my horizons in many areas, especially BPF. Thanks! P.S. This certainly is not a book NOT for absolute beginners.

Few technical books really provide sufficient coverage of the subject material - but Brendan Gregg's 2nd Edition of Systems Performance, Enterprise and the Cloud - is a masterpiece. Within this tome are the techniques that will teach you deep knowledge. When you have synthesized the wisdom within these pages - you will have achieved a level of mastery that will distinguish you for the rest of your career. This book should be on the shelf of every developer, tester, engineer, architect.

Well, this is an amazing book! I strongly recommend it to all IT professionals involved in application development, SRE, sysadmin work, and system performance—and even beyond! 10/10!

Deep understanding, great book

It's fairly easy to pick up on the basics of analyzing system performance, and is one of the skills systems administrators learn early on. You shouldn't confuse that level of knowledge with the one you will have after reading through and learning from this book. Brendan's work in the first, and now greatly expanded and updated, second edition is the gold standard when it comes to moving beyond that novice understanding, and will take you well past an intermediate level knowledge into an advanced practitioner. I make daily use of the things I've learned from both editions. If you have the first edition, the new and refined content here is well worth it (and pick up his BPF book as well if you haven't already!). If you don't own the first edition, it doesn't matter if you're a systems administrator, software developer, devops engineer, or performance engineer with many years of experience - there will be things in here that you did not know and can learn from.

This book has been amazing. I have not read it all the way to the end just quite yet, there's a lot of information to unpack. There is quite a bit of repetition in there which I actually find nice because it refreshes the context but also great as a reference. This has helped me understand quite a bit more about raw performance tuning and how to approach it smartly. I think the methodology was a big help. The tools have also been helpful as its helped me fill in a lot of my blind spots in technology. Again, this is a great book. I'm a general software engineer and this has helped me gain insight into places I don't always get to work directly in. If you're trying to better understand systems in general, I think this is a great book for that too.

I have started working with ebpf and performance related to Kubernetes clusters. The way the chapters flow is like a Jedi Master defining methodologies to cloud computing. The depth and definitions influence my daily work. The is the book along with EPBF Performance that are on my desk, and I consult them on a daily basis.

I have an O'Reilly subscription so technically didn't need to buy this, but it's a quality book filled with fantastic insights, I highly recommend it to anyone interested in systems performance.

Very good book written by someone that has a big knowledge in Performance.

You cannot possibly overestimate importance of having such a book on the table of any serious Linux admin. A wealth of knowledge, and very systematic too.

I just finished reading the book. Read about 85% of the content from cover to cover. I didn't know that there were these kinds of tools and methods available for investigating compute systems and performance. Absolutely critical book to have if you have to spend time analyzing overall performance of compute systems in almost any capacity. The focus is on compute systems with operating systems. If you want advice on how to write optimized assembly for integrated controlers/PLCs or re-organize your C/C++ code to optimize L1/L2 caching accesses etc., you might find the book of limited value but there are still relevant bits in there about benchmarking and other concerns. I am the target audience for this book; my work is on a mix of servers/workstations/cloud compute for simulation and tools that make heavy use of compute power. I am lucky enough to be in on the whole cycle and get hands-on from requirements gathering and design stage to decommissioning and repurposing hardware. This book is already paying dividends in my thought process and approaches.

Conteúdo extraordinário!