14 ms·

The Goal. It's a book from the 1980's about operational management. In particular, it focuses on physical manufacturing. It's in the form of a fictitious pers

The Goal. It's a book from the 1980's about operational management. In particular, it focuses on physical manufacturing. It's in the form of a fictitious personal tale. A plant manager struggles to save his plant from closure, and his marriage from falling apart. The lesson is about lean management. My main takeaway is the realization that when we try to optimize something, we focus on how to do something more efficiently. What more often is a problem is that people and processes are blocked from doing work. They spend a lot of time waiting and doing nothing. Focusing on reducing waits will produce better results than focusing on doing the work faster. Of course reducing the waits might mean doing some targeted piece of work faster. It could also mean doing better scheduling or focusing on other resource contention. Recently I used this mindset to optimize a legacy DB struggling under the weight of a hodgepodge of unmaintained code. It worked wonderfully. Instead of fixing the slowest queries, focused on fixing the ones that block the most often. The result was that the DB was able to handle the workload after all.

gjadi 3y ago
For those interested on this topic but would like a lighter approach, there is now a graphic novel: The Goal A Business Graphic Novel.
michaelhoney 3y ago
Similar theme, not great literature but still a good, quick read for people who work in tech: The Phoenix Project: https://itrevolution.com/product/the-phoenix-project/ https://itrevolution.com/product/the-phoenix-project/
tga 3y ago
The Phoenix Project is basically the same thing, but about software. I also found it more enjoyable to read. https://itrevolution.com/product/the-phoenix-project/ https://itrevolution.com/product/the-phoenix-project/
- abbadadda 3y ago
  The Phoenix Project was in fact inspired by The Goal: > “The Phoenix Project”, a popular business fable about DevOps, is an adaptation of “The Goal” by Dr. Eliyahu Goldratt. They are virtually the same book. The twist it takes on “The Goal” is that instead of focusing on manufacturing, the “The Phoenix Project” focuses on IT. https://sobination.com/2017/08/08/the-goal-and-the-phoenix-project/ https://sobination.com/2017/08/08/the-goal-and-the-phoenix-p...
john-radio 3y ago
Wait, so how do you approach evaluating that for a server that manages a database?
- Cthulhu_ 3y ago
  Isn't most performance problems down to the CPU waiting for something? (network - disk - memory - cache)
  - dsr_ 3y ago
    And almost all the rest are due to the CPU doing unnecessary work.
  - koliber 3y ago
    Sometimes it's the CPU waiting. Sometimes there is a connection open to the DB which acquired a lot of locks during a long-running transaction. The DB is processing this transaction on one DB connection, and many other DB connections are blocked because they can not access certain tables, pages, or rows. The other sessions are literally doing nothing and are waiting. Potential ways to improve it: - Reducing the scope of transactions - committing more often within the transaction - optimizing just the most impactful queries which contribute to the blocking session's runtime (not ones that don't block at all) - loosening the transaction isolation level
- koliber 3y ago
  Previous attempts looked at the heaviest query in terms of processing time, IO, or execution count. Those certainly slowed the server down, but the really issue were application freezes of many seconds and even minutes. Fixing those unblocked the server. The heavy queries are still there, but no one cares. More technically, before, people focused on individual queries with the hope of finding the one that is slowing down the server the most. When we instead focused on sessions, and which ones block others due to locks and DB transactions, we were able to fix the root causes. I am simplifying a bit, but that is the jist.
clumsysmurf 3y ago
Eli's daughter also wrote a more modern take: "Goldratt's Rules of Flow" https://www.amazon.com/gp/product/B0B9VH7TTZ https://www.amazon.com/gp/product/B0B9VH7TTZ
anymouse123456 3y ago
Great recommendation! The book is mainly credited with introducing the Theory of Constraints [1], which asserts that any manageable system's throughput is limited by exactly one bottleneck. He goes on to encourage us to instantiate operating (and thinking) processes whereby we repeatedly discover and mitigate this bottleneck as each mitigation may result in a migration of the constraint. The main observation is that most of us are inclined and almost always incentivized to focus on local optima (our area, our lane, our discipline) and that this striving is a great way to destroy any given system, or at best, do nothing useful. Any small experience in the world of business will illustrate this quite clearly, IME. The sequel to The Goal is also quite good and is called, "It's Not Luck." This book takes the same characters into a larger organization where the constraints are in how their products are marketed, rather than produced in order to more fully demonstrate how the Theory of Constraints is not limited to activities on a factory floor. As mentioned elsewhere, the narrative and writing style can be distracting, but the concepts are timeless and extremely powerful. I have re-read or re-listened to both books at least once every year or two since the early aughts and I learn something new and relevant with each pass. [1] https://en.wikipedia.org/wiki/Theory_of_constraints https://en.wikipedia.org/wiki/Theory_of_constraints [2] https://www.goodreads.com/book/show/157385.It_s_Not_Luck https://www.goodreads.com/book/show/157385.It_s_Not_Luck
- shanusmagnus 3y ago
  I tried to read The Goal and bounced off due in part to the aforementioned writing style, which seemed super trite and straw-mannish. Does the fact that you keep re-reading the books suggests that you haven't found a better way to re-approach that material?
  - koliber 3y ago
    I listened to it as an audio book. It was possible to ignore the style. It simplifies situations and trivializes the road to success bit. However, that seems necessary to keep the book clear, to the point, and of manageable size.
  - anymouse123456 3y ago
kimchidude 3y ago
I really enjoyed that book and how it attempts to simplify the nature of common supply chain constraints. You probably know this already, but if not: Eli Goldratt (the guy that wrote that book) wrote an equally brilliant academic paper on the same subject - I think the title was ‘Standing on the shoulders of giants’. Just passing it on in the event you’re looking for a good follow-up read.
- koliber 3y ago
  Yes! The paper is in the appendix of the book. Where the book is a story and does not mention Lean Management, Toyota, and Taiichi Ohno, "Standing on the shoulders of giants" is a clear description of TPS (Toyota Production System). It's awesome how the two varied styles of writing describe the same "thing". Where the "thing" is "how to operate efficiently".

gjadi 3y ago

For those interested on this topic but would like a lighter approach, there is now a graphic novel: The Goal A Business Graphic Novel.

michaelhoney 3y ago

Similar theme, not great literature but still a good, quick read for people who work in tech: The Phoenix Project: https://itrevolution.com/product/the-phoenix-project/ https://itrevolution.com/product/the-phoenix-project/

tga 3y ago

The Phoenix Project is basically the same thing, but about software. I also found it more enjoyable to read. https://itrevolution.com/product/the-phoenix-project/ https://itrevolution.com/product/the-phoenix-project/

abbadadda 3y ago
The Phoenix Project was in fact inspired by The Goal: > “The Phoenix Project”, a popular business fable about DevOps, is an adaptation of “The Goal” by Dr. Eliyahu Goldratt. They are virtually the same book. The twist it takes on “The Goal” is that instead of focusing on manufacturing, the “The Phoenix Project” focuses on IT. https://sobination.com/2017/08/08/the-goal-and-the-phoenix-project/ https://sobination.com/2017/08/08/the-goal-and-the-phoenix-p...

john-radio 3y ago

Wait, so how do you approach evaluating that for a server that manages a database?

Cthulhu_ 3y ago
Isn't most performance problems down to the CPU waiting for something? (network - disk - memory - cache)
- dsr_ 3y ago
  And almost all the rest are due to the CPU doing unnecessary work.
- koliber 3y ago
  Sometimes it's the CPU waiting. Sometimes there is a connection open to the DB which acquired a lot of locks during a long-running transaction. The DB is processing this transaction on one DB connection, and many other DB connections are blocked because they can not access certain tables, pages, or rows. The other sessions are literally doing nothing and are waiting. Potential ways to improve it: - Reducing the scope of transactions - committing more often within the transaction - optimizing just the most impactful queries which contribute to the blocking session's runtime (not ones that don't block at all) - loosening the transaction isolation level
koliber 3y ago
Previous attempts looked at the heaviest query in terms of processing time, IO, or execution count. Those certainly slowed the server down, but the really issue were application freezes of many seconds and even minutes. Fixing those unblocked the server. The heavy queries are still there, but no one cares. More technically, before, people focused on individual queries with the hope of finding the one that is slowing down the server the most. When we instead focused on sessions, and which ones block others due to locks and DB transactions, we were able to fix the root causes. I am simplifying a bit, but that is the jist.

clumsysmurf 3y ago

Eli's daughter also wrote a more modern take: "Goldratt's Rules of Flow" https://www.amazon.com/gp/product/B0B9VH7TTZ https://www.amazon.com/gp/product/B0B9VH7TTZ

anymouse123456 3y ago

Great recommendation! The book is mainly credited with introducing the Theory of Constraints [1], which asserts that any manageable system's throughput is limited by exactly one bottleneck. He goes on to encourage us to instantiate operating (and thinking) processes whereby we repeatedly discover and mitigate this bottleneck as each mitigation may result in a migration of the constraint. The main observation is that most of us are inclined and almost always incentivized to focus on local optima (our area, our lane, our discipline) and that this striving is a great way to destroy any given system, or at best, do nothing useful. Any small experience in the world of business will illustrate this quite clearly, IME. The sequel to The Goal is also quite good and is called, "It's Not Luck." This book takes the same characters into a larger organization where the constraints are in how their products are marketed, rather than produced in order to more fully demonstrate how the Theory of Constraints is not limited to activities on a factory floor. As mentioned elsewhere, the narrative and writing style can be distracting, but the concepts are timeless and extremely powerful. I have re-read or re-listened to both books at least once every year or two since the early aughts and I learn something new and relevant with each pass. [1] https://en.wikipedia.org/wiki/Theory_of_constraints https://en.wikipedia.org/wiki/Theory_of_constraints [2] https://www.goodreads.com/book/show/157385.It_s_Not_Luck https://www.goodreads.com/book/show/157385.It_s_Not_Luck

shanusmagnus 3y ago
I tried to read The Goal and bounced off due in part to the aforementioned writing style, which seemed super trite and straw-mannish. Does the fact that you keep re-reading the books suggests that you haven't found a better way to re-approach that material?
- koliber 3y ago
  I listened to it as an audio book. It was possible to ignore the style. It simplifies situations and trivializes the road to success bit. However, that seems necessary to keep the book clear, to the point, and of manageable size.
- anymouse123456 3y ago

kimchidude 3y ago

I really enjoyed that book and how it attempts to simplify the nature of common supply chain constraints. You probably know this already, but if not: Eli Goldratt (the guy that wrote that book) wrote an equally brilliant academic paper on the same subject - I think the title was ‘Standing on the shoulders of giants’. Just passing it on in the event you’re looking for a good follow-up read.

koliber 3y ago
Yes! The paper is in the appendix of the book. Where the book is a story and does not mention Lean Management, Toyota, and Taiichi Ohno, "Standing on the shoulders of giants" is a clear description of TPS (Toyota Production System). It's awesome how the two varied styles of writing describe the same "thing". Where the "thing" is "how to operate efficiently".