What a surprisingly busy week.
Spent quite a lot of time tracking down a linking problem on Itanium. It appears that GNU ld 2.14 has some serious deficiencies. GNU ld 2.17 (the current stable release) does fix the problem. Unfortunately, the last supported version of GNU ld on RedHat AS 3 for Itaniam is version 2.14.90.0.4.20030529 - which clearly cannot contain the fix which was written in November of 2003. I suppose what to do is now a decision to be made by someone else.
Also spent some time with Federated - Bulk insert support within the storage engine and also how to make Federated play better with transactions and network overhead. Have to put together a new WorkLog entry if I cannot find a relevant one. I need to work on writing better changeset comments. Ingo says, and I have to admit that I agree a bit, that my comments assume that the reader is familiar with the code and bug in question. I need to make them a little more verbose so that anyone can work out what is being fixed without them having to dig through the bug database.
I had a few interesting chats/email discussions to discover that perhaps why there are relatively few complaints about Federated is because while it works for simple cases, there are lot of areas where it doesn't work for users and so it quickly gets abandoned and no bug report is filed, which is a shame as Federated can be a really useful tool.
I have removed the my Sparc and Power Falcon trees. I don't really have time to maintain them anymore: It has been nearly two weeks since I last merged, built and tested them and with all the new tablespace work going in to the tree, future merges are likely to be problematic. Doing this will liberate a couple of hours a week so that I can examine other interesting stuff or perhaps get some more bugs fixed.
08 June 2007
30 May 2007
Book Review: Understanding MySQL Internals.
I actually received this book about a month ago but from necessity, I had to put it to one side and deal with more important issues. It is not a big book, only 235 pages, but it has an easy to read conversational style, unlike similarly titled technical books on the internals of other software.
If you're looking for a book about database theory, this book is not for you. It conveys to the reader a good overview as how MySQL functions without delving into the theories as to how the algorithms work or the rationale behind their design. The book does a good job of at least giving a drive-by look at most of the significant components of MySQL. libmysys, MySQL's platform library which allows MySQL to work on a very wide range of operating systems is mentioned only under the subsection titles 'Core API' and 'Utility API' and only lists about 30 of the functions available: libmysys provides the important glue between MySQL and the operating system, comparable with Netscape's NSPR or Apache's APR. The only omission within this book is information of MySQL's string-handling and character-set functions and data structures.
As the author acknowledged in his introduction, the bleeding-edge of MySQL development does move quite fast - so for a reader to make use of this book, I would recommend using the MySQL 4.1 or 5.0 source code for reference. There are many significant changes within the 5.1 beta source code which are not covered - which is perfectly reasonable as 5.1 is still in beta. I do hope that Sasha Pachev has a "2nd Edition" in the works which will cover 5.1 when it is finally released as stable and the plug-ins infrastructure as that would lie the greatest avenue for users and 3rd party ISVs to develop extensions for MySQL. There are also several exciting new components, such as the Falcon storage engine, which would be worthy of mention and perhaps an overview of their operation.
This book would be valuable to anyone unfamiliar with the MySQL source code and provides a foundation which can be built upon. It is informative and does not dazzle the reader with techo-jargon and theories.
If you're interested in this book, it is published by O'Reilly.
http://www.oreilly.com/catalog/9780596009571/index.html
If you're looking for a book about database theory, this book is not for you. It conveys to the reader a good overview as how MySQL functions without delving into the theories as to how the algorithms work or the rationale behind their design. The book does a good job of at least giving a drive-by look at most of the significant components of MySQL. libmysys, MySQL's platform library which allows MySQL to work on a very wide range of operating systems is mentioned only under the subsection titles 'Core API' and 'Utility API' and only lists about 30 of the functions available: libmysys provides the important glue between MySQL and the operating system, comparable with Netscape's NSPR or Apache's APR. The only omission within this book is information of MySQL's string-handling and character-set functions and data structures.
As the author acknowledged in his introduction, the bleeding-edge of MySQL development does move quite fast - so for a reader to make use of this book, I would recommend using the MySQL 4.1 or 5.0 source code for reference. There are many significant changes within the 5.1 beta source code which are not covered - which is perfectly reasonable as 5.1 is still in beta. I do hope that Sasha Pachev has a "2nd Edition" in the works which will cover 5.1 when it is finally released as stable and the plug-ins infrastructure as that would lie the greatest avenue for users and 3rd party ISVs to develop extensions for MySQL. There are also several exciting new components, such as the Falcon storage engine, which would be worthy of mention and perhaps an overview of their operation.
This book would be valuable to anyone unfamiliar with the MySQL source code and provides a foundation which can be built upon. It is informative and does not dazzle the reader with techo-jargon and theories.
If you're interested in this book, it is published by O'Reilly.
http://www.oreilly.com/catalog/9780596009571/index.html
Subscribe to:
Posts (Atom)
