This thread was mentioned in a performance sub-group posting. Give it a read.
Back? It means, so far as I can see, that PG is toast. It will fall down to being the cheap and dirty alternative to MySql, which even has, at least two, multi-threaded engines. DB2 switched it's *nix engine to threads from processes with release 9.5. Oracle claims it for releases going back to 7 (I haven't tried to determine which parts or applications; Larry has bought so many tchochtkes over the years...). SQL Server is threaded.
Given that cpu's are breeding threads faster than cores, PG will fall into irrelevance. Too bad, it's kind of a nice database.
Despite what the reactionaries assert, the world is not flat nor is it hierarchical. The world is relational. Infinite granularity yields infinite complexity.
Vaccinated ≠ Not Infectious
$536,800,000 MARF™(party like it's 1829)
-- New York Slammers, so far/2024 [GA and The Feds still to come]
-- New York Slammers, so far/2024 [GA and The Feds still to come]
Covid-19 has killed at least 1,123,836 people (as of 20 March 2023, final update)
Scientists aren't vocal enough about science. There are large groups of people who think scientists are all frauds and who don't
believe in science, and they're being cultured by some of our far right-wing politicians, religious leaders and community leaders. -- Drew Weissman/2022
To date, Microsoft is stating that organizations testing In-Memory OLTP have
seen transaction speeds improve by up to 30 times compared to past
performance, with the best performance gains achieved when the business
logic resides in the database and not in the applications.
-- Jonathan Watts/2015 [my emphasis]
-- Jonathan Watts/2015 [my emphasis]
I have had to explain and re-explain and re-explain and re-explain, you know, how relational databases work, what is an eigenvector,
what is dimensionality reduction.
-- Christopher Wylie/2018
... but Flash-based storage has such a different performance profile from rotating media, that I suspect that it will end up having
a large impact on filesystem design. Right now, most filesystems tend to be designed with the latencies of rotating media in mind.
-- Linus Torvalds/2007
-- Chris Date/2009
This week's thought
[The intelligence community] argued that [the restrictive amendment] would cripple the program because they typically use it before there is enough evidence to meet a standard of probable cause for a warrant, like early in investigations when they are trying to learn more about a phone number or an email account found to be in contact with a suspected foreign spy or terrorist. -- Charlie Savage/2024 [IOW, let's catch the thief before the horse is out of the barn - the MARF™(party like it's 1829) traitor's worst nightmare]
In a time of SSD, multi-core/processor, two terabyte memory and Optane App Direct Mode machines, there is no reason not to
build from BCNF data. Time to do what Dr. Codd demonstrated. Technology has finally caught up with the maths.
About
- Robert Young
- Been doing databases, both relational and others, as well as operations research since Hector was a Pup.
Shameless Plug
I know a little something about SQL, DB2 (preferably LUW, z/OS if the money's right), Oracle, SQL Server, Postgres, and database design in general. Mentoring/teaching/consulting on a short term basis, too. And a bit of stat analysis if it's R.
gnuoytr at rcn dot com.
gnuoytr at rcn dot com.
4 comments:
The author of that post apparently doesn't understand that even though postgresql hasn't 'switched to threads', it can still do more than one thing at once. Each process is itself an execution thread. A multi-threaded query planner is perfectly possible in postgresql architecture -- however each one must reside in it's own process and you have to use shared memory instead instead of pthreads and locking. Big whoop. The only thing at stake with a multi threaded planner is optimizing single user tasks which is, while important, a niche optimization. PostgreSQL is for more scalable than mysql for multi-user loads and the gap is increasing.
The author also has not understood that Oracle runs multiple processes just like PostgreSQL does - at least on Linux/Unix platforms. Oracle uses a thread model only on Windows.
http://docs.oracle.com/cd/E11882_01/server.112/e25789/process.htm#i16977
hi
wanna say - oracle use process in unix version , and threads in windows version.
And windows version of oracle is mooost slower than unix version.
I think is it not so clear about threads and process as you say.
If RDBMS uses threads - any way them must use the shared memory, becas multithreaded rdbms use lot than one process anyway(more than one cpu core).
This mean using threads is not a unicorn, and depends of concrete realization.
about ms sql server threads - it's not a feature, it's a windows system restriction - windows processes cant use shared memory like a unux. (and this a reason why oracle do the some).
and, if thread crash - them crash the parent process with all other threads in this process.
For the rdbms with lot of calculations (not only read\write) - it's critical restriction.
Post a Comment