Understanding transactions

On Thu, Jun 03, 2004 at 05:38:21PM -0700, Jonathan Gardner wrote:

Do you know of good resources on learning how transactions are implemented
in PostgreSQL? Should I just peruse the code? Which files are most
relevant?

backend/access/transam/xact.c (high level transaction status) and
backend/utils/time/tqual.c (low level tuple visibility test --- this is
hard to read and understand) and its respective header files are the most
relevant. See also htup.h about "heap tuples," specially heap tuple
headers and the games with Xmin, Xmax, Cmin and Cmax.

Not sure about other literature. Make sure you understand how MVCC
works before reading the code ...

--
Alvaro Herrera (<alvherre[a]dcc.uchile.cl>)
Y dijo Dios: "Que sea Satan�s, para que la gente no me culpe de todo a m�."
"Y que hayan abogados, para que la gente no culpe de todo a Satan�s"

Jonathan Gardner wrote:

Seeing how PITR, nested transactions, and other exciting developments
related to transactions are being developed, I am getting curious about how
PostgreSQL actually implements transactions. Investigating Materialized
Views has led me to look closely at how transactions work and such.

Do you know of good resources on learning how transactions are implemented
in PostgreSQL? Should I just peruse the code? Which files are most
relevant?

Next to diving into the code you might want to dive into the
theory. Chapter 5 of "Concurrency Control and Recovery in
Database Systems" deals with Multi Version Concurrency Control:
http://research.microsoft.com/pubs/ccontrol/

Jochem