|
most sql databases are actually good. at least when you compare with the nosql competition
|
# ? May 19, 2016 03:19 |
|
|
# ? Apr 26, 2024 22:14 |
|
tef posted:schemas and migrations are great when you can do them without downtime yeah that's why I'm that thing
|
# ? May 19, 2016 03:31 |
|
I really don't want to have to learn SQL but I bet I will have to at some point.
|
# ? May 19, 2016 03:49 |
|
sql is great.
|
# ? May 19, 2016 03:52 |
|
tef posted:here's a database riddle: how do you do PRODUCT(column), i.e c1*c2*c3*c4... code:
|
# ? May 19, 2016 04:14 |
|
nice the old fashioned way is to do exp(sum(log(col)))
|
# ? May 19, 2016 04:17 |
|
hahahaha oh man i never would have thought of that, and if i saw it in existing code i wouldn't have understood the purpose either. it's perfect
|
# ? May 19, 2016 04:19 |
|
"but jewkiller, you racist gently caress, surely it must also be easy to make a custom aggregate function in oracle to do products, right?" you bet!http://radino.eu/2010/11/17/product-aggregate-function/ posted:
ENTERPRISE!
|
# ? May 19, 2016 04:27 |
|
qhat posted:for example Debian mysql stack
|
# ? May 19, 2016 06:57 |
|
SpaceAceJase posted:Debian mysql stack don't even say mysql is good tia
|
# ? May 19, 2016 12:37 |
|
my stepdads beer posted:don't even say mysql is good tia the attention i crave
|
# ? May 19, 2016 12:45 |
|
id just go with postgres and centos if i needed a sql
|
# ? May 19, 2016 13:03 |
|
if i needed a full text search to go with it id probably throw in elastic
|
# ? May 19, 2016 13:04 |
|
elasticsearch is more powerful, but postgres full text search is often good enough, and then you don't have to maintain a separate search engine
|
# ? May 19, 2016 17:14 |
|
JewKiller 3000 posted:"but jewkiller, you racist gently caress, surely it must also be easy to make a custom aggregate function in oracle to do products, right?" you bet! i've been reliably informed that you picked the dumbest possible way to illustrate this
|
# ? May 19, 2016 17:16 |
|
go ahead then, show me the nice easy way to make a custom aggregate in oracle i mean that's practically straight out of the manual so idk how reliable your informant is
|
# ? May 19, 2016 17:22 |
|
hey, i didn't read any of this and i'm not going to. just wanted to say that mariadb is really good, but if you can't use that plain old mysql is just fine
|
# ? May 19, 2016 17:27 |
|
many thanks, tiny bug idiot, for blindly contributing your awful phpinions to yet another thread. your gimmick is truly always appreciated
|
# ? May 19, 2016 17:44 |
|
i've only ever implemented a non-trivial mysql production database once and that was and will be the last time
|
# ? May 19, 2016 17:55 |
|
it's un loving believable the amount of hoops you have to jump through to get that DB to work well enough to be considered production ready. but hey i guess if you hate stored procedures and replication because you see those features as just unnecessary and for spergs only, then go loving nuts
|
# ? May 19, 2016 17:58 |
|
sproc out with you're cock out
|
# ? May 19, 2016 18:11 |
|
Tiny Bug Child posted:hey, i didn't read any of this and i'm not going to. just wanted to say that mariadb is really good, but if you can't use that plain old mysql is just fine Don't you outsource your db poo poo to some webhost anyway
|
# ? May 19, 2016 19:20 |
|
qhat posted:it's un loving believable the amount of hoops you have to jump through to get that DB to work well enough to be considered production ready. but hey i guess if you hate stored procedures and replication because you see those features as just unnecessary and for spergs only, then go loving nuts isn't it just?? getting a slave up on a large existing database is a huge pain in the dick that involves special percona perl tools that barely work and non trivial amounts of downtime due to needing a long global lock on the master. or even more downtime if you do it hte official way and shutdown the master and rsync the data the slave
|
# ? May 19, 2016 23:35 |
|
what i learned recently is that text indexes on mssql (all sql maybe?) are basically only starts with or ends with otherwise everything is just a scan so %mystring% is always a scan, who knew? when i did some "optimisation" by adding indexes to common columns i foudn this out because i created a query plan that was 50/50 as to whether it would ultra gently caress the query processor into doing 2 million read ops for a regular query or 10 read ops, can you guess which version ran in testing vs. production?
|
# ? May 19, 2016 23:56 |
|
my stepdads beer posted:isn't it just?? getting a slave up on a large existing database is a huge pain in the dick that involves special percona perl tools that barely work and non trivial amounts of downtime due to needing a long global lock on the master. or even more downtime if you do it hte official way and shutdown the master and rsync the data the slave Started a new slave today with 300Gb of MySQL data using Percona xtrabackup and it worked great with no downtime. Recommended 10/10 tool.
|
# ? May 20, 2016 00:02 |
|
300gb. bless.
|
# ? May 20, 2016 00:26 |
|
JewKiller 3000 posted:elasticsearch is more powerful, but postgres full text search is often good enough, and then you don't have to maintain a separate search engine oh that's nice
|
# ? May 20, 2016 01:26 |
|
my stepdads beer posted:300gb. bless. that's a lot of butts
|
# ? May 20, 2016 01:31 |
|
should i make a database paper effort post with some of the random poo poo i've been reading? http://www.msr-waypoint.com/pubs/178758/bw-tree-icde2013-final.pdf http://db.disi.unitn.eu/pages/VLDBProgram/pdf/research/p853-levandoski.pdf like this pair, bw-tree and llama log strructured storage or these old ones http://daslab.seas.harvard.edu/reading-group/papers/kung.pdf http://csis.pace.edu/~marchese/CS865/Papers/p180-thomas.pdf
|
# ? May 20, 2016 19:47 |
|
are you actually implementing your own database?
|
# ? May 20, 2016 19:53 |
|
tef posted:should i make a database paper effort post with some of the random poo poo i've been reading? yes.
|
# ? May 20, 2016 19:59 |
|
|
# ? May 20, 2016 20:02 |
|
no. i don't need more poo poo on my reading list
|
# ? May 20, 2016 20:10 |
|
YeOldeButchere posted:no. i don't need more poo poo on my reading list if you don;t want to read teffort posts get the gently caress out of yospos forever
|
# ? May 20, 2016 20:13 |
|
Share Bear posted:are you actually implementing your own database? i've pissed around with some mvcc toys and made a radix trie, but never really went into wal/recovery (the hard bit) i'd like to for fun and to learn something but it is a lot of effort, but having a vague idea about that toy helps me find different papers to read
|
# ? May 20, 2016 20:17 |
|
that's cool. the older papers are especially helpful for understanding what may be going on under the hood as well. old papers also seem remarkably easier to read than the newer ones
|
# ? May 20, 2016 20:21 |
|
Share Bear posted:that's cool. the older papers are especially helpful for understanding what may be going on under the hood as well. a lot of the harder ideas in old papers are often more common today a lot of the modern papers are very clever optimisations or applications of earlier principles there's still some the other way around though, but in general the older it is the more likely you've been exposed to the harder concepts contained and then it's down to deciphering the notation and jargon
|
# ? May 20, 2016 20:34 |
|
that and time tends to make it easier to filter out the god awful ones b/c no-one cites them
|
# ? May 20, 2016 20:35 |
|
time for me to shill for "Readings in Database Systems" which is pretty much a collection of classic or otherwise interesting db papers. it's pretty good
|
# ? May 20, 2016 20:43 |
|
|
# ? Apr 26, 2024 22:14 |
|
Captain Foo posted:read teffort posts
|
# ? May 21, 2016 01:36 |