- developers - lists.mariadb.org

Re: [Maria-developers] Installing jemalloc on build VMs
by Kristian Nielsen 26 Apr '13

26 Apr '13

Daniel Bartholomew <dbart(a)mariadb.org> writes: > I'm getting ready to install jemalloc on the buildslaves for TODO-459 - > https://mariadb.atlassian.net/browse/TODO-459 , and wanted to know if > there was a preferred way to do it. Since we want to use a specific > version of jemalloc (3.1.0), I figured I would build from source on all > of the build machines. Has this been thought through properly with respect to packaging? My guess is it has not. For .debs, which are what I know about, it is generally expected that one can rebuild from `apt-get source mariadb-server`. This requires proper build dependencies. It doesn't work to install random stuff on random machines. It does not appear that jemalloc 3.1 is available on most .deb distros we support. Wheezy seems to have 3.0.0. It is said that tokutek "requires" jemalloc - maybe it would be better to fix tokutek to work with standard libraries? The proper way to produce .debs with custom jemalloc requires creating new custom packages for libjemalloc and including them in our repos. - Kristian.

8 11

Re: [Maria-developers] MariaDB and jemalloc
by Axel Schwenke 24 Apr '13

24 Apr '13

Hi Sanja, Oleksandr Byelkin wrote: > 24.04.2013 19:00, Axel Schwenke пишет: > [skip] >> >> OLTP ro transactions per second relative to glibc malloc >> ---------------------------------------------------------------------- >> Threads jemalloc-3.3.1 glibc malloc tcmalloc jemalloc-3.1.0 >> 4 +3.2% +0.0% +11.6% +61.9% >> 8 +0.9% +0.0% -2.3% +8.0% >> 16 +9.6% +0.0% +4.7% +12.9% >> 32 +1.8% +0.0% +0.9% +5.8% >> 64 -5.1% +0.0% +2.1% +5.4% >> 128 -4.8% +0.0% +0.2% +4.5% >> 256 -4.2% +0.0% +0.7% +5.0% >> 512 -13.3% +0.0% +1.8% +4.1% > > Strange that results jumps up and down when percona had quite smooth > one. Is it possible that computer was doing something else? or we need > more tests to take average? It's not so "jumpy" when you look at the absolute numbers. The difference to Percona is, that they tested their own server, where MariaDB tries to minimize the number of malloc() calls. I wouldn't give too much on the numbers for 4 threads, since this is the buffer pool warmup phase. Also I've run each test for only 100 seconds to get results fast. Re "jumpy" - see the attached pdf with the scatter plot. Each point represents the tps from a 5 second interval. Numbers are in fact rather smooth, except for 512 threads. The main takeaway is, that I can reproduce the regression in jemalloc 3.3. With TokuDB there is no way around jemalloc anyway. XL

2 1

[Maria-developers] MariaDB and jemalloc
by Axel Schwenke 24 Apr '13

24 Apr '13

Hi, Monty asked me to do a quick check how MariaDB performs with jemalloc. This is in the light of the upcoming integration of TokuDB into the MariaDB tree. TokuDB requires jemalloc. While I did use tcmalloc in the connect speed benchmarks with good results, I did no use jemalloc before. This article from Percona http://www.mysqlperformanceblog.com/2013/03/08/mysql-performance-impact-of-… suggests that jemalloc and tcmalloc perform very comparable, but also that jemalloc shows a regression in 3.2.0 and later. The benchmark I used is sysbench OLTP ro, 16 tables, data fitting into buffer pool. I used the new 64 (virtual, 32 real) core machine and compared 4 different malloc implementations as below: OLTP ro transactions per second relative to glibc malloc ---------------------------------------------------------------------- Threads jemalloc-3.3.1 glibc malloc tcmalloc jemalloc-3.1.0 4 +3.2% +0.0% +11.6% +61.9% 8 +0.9% +0.0% -2.3% +8.0% 16 +9.6% +0.0% +4.7% +12.9% 32 +1.8% +0.0% +0.9% +5.8% 64 -5.1% +0.0% +2.1% +5.4% 128 -4.8% +0.0% +0.2% +4.5% 256 -4.2% +0.0% +0.7% +5.0% 512 -13.3% +0.0% +1.8% +4.1% (more data in attached spread sheet) Observations: - the effect is rather limited - jemalloc 3.1.0 clearly wins - jemalloc 3.3.1 performs bad at higher concurrency Conclusion It's fine to build MariaDB with jemalloc in the future. However we mustn't use jemalloc-3.3.1, but rather jemalloc-3.1.0. XL

1 0

[Maria-developers] Fwd: Fwd: gsoc13
by Samarth Mathur 24 Apr '13

24 Apr '13

Hi! I have a few doubts. Going through the internet, I've come across two major implementations of LDAP protocol The openLDAP, and the libpam_ldap library. Both have a detailed information of client side implementation. However, openLDAP gives the slapd which is a standalone server daemon, whereas from the libpam_ldap library, I am able to gather the existence of an inbuilt ldap authentication system. Am I correct about that? Which do you think will be useful to implement as a plugin for mariadb? Thanks!

2 2

[Maria-developers] log analyzer
by Roberto Spadim 23 Apr '13

23 Apr '13

Hi guys, I was looking for something that give me the 'prepared statement' of a SQL query for example: statment = SELECT * FROM table WHERE field="value" prepared statement = SELECT * FROM table WHERE field=? why do this? well my idea is build a log analyzer and get commong queries since some programs don't use native prepared statement, reading the slow query log, or query log only will give me many different queries what i want is: 1)read queries log 2)get queries 3)get the 'prepared statement' of the query 4)create a table with prepared statement / unique id (maybe md5 of (database name + prepared statement) ) 5)create statistics using the unique id of prepared statement and check if the same 'kind' of queries are running with a mean time, or have some low and high times (for example) or getting information if some index is being used or not, well here i have many ideas but it's not the 'problem' for now the problem is the item (3), i have a query and i need the 'prepared statement' of this query, even the query was not executed with prepared statements this should work for INSERT, DELETE, UPDATE, SELECT, CREATE TABLE and others queries too (for now SELECT is enought, but in future others kinds of query will be nice) anyone know something that could do this job? maybe part of the mariadb query parser could/should be used? i was thinking about a command that could be executed in mariadb, today we have the EXPLAIN, we could have PARSE or other name? simplifying the problem: read the query, parse it and change constant values with "?" willcard thanks! any help is welcome -- Roberto Spadim

1 2

[Maria-developers] MariaDB 10.0.2-alpha release prep
by Daniel Bartholomew 23 Apr '13

23 Apr '13

I've begun prepping MariaDB 10.0.2-alpha for release. Draft changelog and release notes are here: - https://kb.askmonty.org/en/mariadb-1002-release-notes/ - https://kb.askmonty.org/en/mariadb-1002-changelog/ In the draft release notes I've listed MDEV-26 (Global Transaction ID), and the CONNECT handler. Please let me know of other features or fixes I should highlight ASAP. Thanks. -- Daniel Bartholomew Google+ - http://gplus.to/dbart Twitter - http://twitter.com/daniel_bart MariaDB: An Enhanced Drop-in Replacement for MySQL Website - http://mariadb.org Twitter - http://twitter.com/mariadb Google+ - http://google.com/+mariadb Facebook - http://fb.com/MariaDB.dbms Knowledgebase - http://kb.askmonty.org Monty Program - http://montyprogram.com

2 1

[Maria-developers] ALTER TABLE xxx ENGINE=aria, progress only works in copying stage
by Roberto Spadim 21 Apr '13

21 Apr '13

hi guy, i was changing myisam to aria on some big tables (15gb) and show processlist stop reset the progress at 16% the operation started with 'copying to tmp table', and i saw 1%, 2% ... 16% now the table is doing "repair by sorting" and progress is 0% the point is... could the processlist show, at least, the last progress percentage? without get back to 0%? or maybe a column of current operation progress and another of total process operation? does 'repair by sorting' have progress algoritm implemented? i'm running mariadb 10.0.1 on linux 64bits thanks guys -- Roberto Spadim

1 0

[Maria-developers] performance monitoring
by MARK CALLAGHAN 21 Apr '13

21 Apr '13

After reading this I wonder if there is an opportunity for SkySQL and Percona to offer "performance tuning for performance schema" classes. I don't think performance monitoring should be this hard. What are the plans for MariaDB moving forward -- more user_stats/table_stats/index_stats and maybe statement_stats or integrate with the PS? http://dev.mysql.com/doc/refman/5.6/en/performance-schema.html http://marcalff.blogspot.com/2013/04/on-configuring-performance-schema.html -- Mark Callaghan mdcallag(a)gmail.com

2 1

Re: [Maria-developers] Fwd: gsoc13
by Sergei Golubchik 21 Apr '13

21 Apr '13

Hi, Samarth! On Apr 20, Samarth Mathur wrote: > "See also the authentication plugin examples in plugin/auth/." > > I couldn't locate this folder in the source code (mariadb 10). > Does it not exist there? No, but there are mariadb-10.0 $ ls -dl plugin/auth* drwx------ 3 serg serg 4096 Apr 20 10:13 plugin/auth_dialog drwx------ 3 serg serg 4096 Apr 20 10:13 plugin/auth_examples drwx------ 5 serg serg 4096 Apr 20 10:13 plugin/auth_pam drwx------ 3 serg serg 4096 Apr 20 10:13 plugin/auth_socket make your choice. Regards, Sergei

1 0

[Maria-developers] Aria engine integrity
by Roberto Spadim 21 Apr '13

21 Apr '13

Hi guys, another doubt about mariadb, now about aria... i'm using aria since myisam is a bit unsecure and have a slow recovery/repair process, aria storage is more secure my doubt is... could i use aria engine over a ext2 file system and have a 'security' near a jornaled file system (xfs for example) + myisam? or maybe xfs+aria? what's better? just one repair, from aria engine, or two repairs, one from file system and other from aria? thanks guys -- Roberto Spadim

3 3