- developers - lists.mariadb.org

[Maria-developers] WL#70 Deleted (by Serg): Fix table_cache negative scalability
by worklog-noreply＠askmonty.org 28 Jun '10

28 Jun '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Fix table_cache negative scalability CREATION DATE..: Fri, 18 Dec 2009, 16:30 SUPERVISOR.....: Bothorsen IMPLEMENTOR....: COPIES TO......: CATEGORY.......: Server-Sprint TASK ID........: 70 (http://askmonty.org/worklog/?tid=70) VERSION........: WorkLog-3.4 STATUS.........: Un-Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: DESCRIPTION: Fix the problem described in this blog entry: http://www.mysqlperformanceblog.com/2009/11/16/table_cache-negative-scalabi… You can read the blog, or the text below. --- quoted text --- November 16, 2009 table_cache negative scalability Posted by peter | Vote on Planet MySQL Couple of months ago there was a post by FreshBooks on getting great performance improvements by lowering table_cache variable. So I decided to investigate what is really happening here. The common sense approach to tuning caches is to get them as large as you can if you have enough resources (such as memory). With MySQL common sense however does not always works weve seen performance issues with large query_cache_size also sort_buffer_size and read_buffer_size may not give you better performance if you increase them. I found this also applies to some other buffers. Even though having previous experience of surprised behavior I did not expect such a table_cache issue the LRU for cache management is classics and there are scalable algorithms to deal with it. I would expect Monty to implement one of them. To do the test I have created 100.000 empty tables containing single integer column and no indexes and when ran SELECT * FROM tableN in the loop. Each table in such case is accessed only once and on any but first run each access would require table replacement in table cache based on LRU logic. MySQL Sandbox helped me to test this with different servers easily. I did test on CentOS 5.3, Xeon E5405, 16GB RAM and EXT3 file system on the SATA hard drive. MySQL 5.0.85 Created 100.000 tables in around 3min 40 sec which is about 450 tables/sec This indicates the fsync is lying on this test system as default sync_frm option is used. With default table_cache=64 accessing all tables take 12 sec which is almost 8500 tables/sec which is a great speed. We can note significant writes to the disk during this read-only benchmark. Why ? Because for MyISAM tables table header has to be modified each time the table is opened. In this case the performance was so great because all 100.000 tables data (first block of index) was placed close by on disk as well as fully cached which made updates to headers very slow. In the production systems with table headers not in OS cache you often will see significantly low numbers 100 or less. With significantly larger table_cache=16384 (and appropriately adjusted number of open files) the same operation takes 660 seconds which is 151 tables/sec which is around 50 times slower. Wow. This is the slow down. We can see the load becomes very CPU bound in this case and it looks like some of the table_cache algorithms do not scale well. The absolute numbers are also very interesting 151 tables/sec is not that bad if you look at it as an absolute number. So if you tune table cache is normal case and is able to bring down your miss rate (opened_tables) to 10/sec or less by using large table_cache you should do so. However if you have so many tables you still see 100+ misses/sec while your data (at least table headers) is well cached so the cost of table cache miss is not very high, you may be better of with significantly reduced table cache size. The next step for me was to see if the problem was fixed in MySQL 5.1 in this version table_cache was significantly redone and split in table_open_cache and table_definition_cache and I assumed the behavior may be different as well. MySQL 5.1.40 I started testing with default table_open_cache=64 and table_definition_cache=256 the read took about 12 seconds very close to MySQL 5.0.85. As I increased table_definition_cache to 16384 result remained the same so this variable is not causing the bottleneck. However increasing table_open_cache to 16384 causes scan to take about 780 sec which is a bit worse than MySQL 5.0.85. So the problem is not fixed in MySQL 5.1, lets see how MySQL 5.4 behaves. MySQL 5.4.2 MySQL 5.4.2 has higher default table_open_cache so I took it down to 64 so we can compare apples to apples. It performs same as MySQL 5.0 and MySQL 5.1 with small table cache. With table_open_cache increased to 16384 the test took 750 seconds so the problem exists in MySQL 5.4 as well. So the problem is real and it is not fixed even in Performance focused MySQL 5.4. As we can see large table_cache (or table_open_cache_ values indeed can cause significant performance problems. Interesting enough Innodb has a very similar task of managing its own cache of file descriptors (set by innodb_open_files) As the time allows I should test if Heikki knows how to implement LRU properly so it does not have problem with large number. Well see. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v4.0.0)

1 0

[Maria-developers] WL#72 Deleted (by Serg): Fix table_cache negative scalability
by worklog-noreply＠askmonty.org 28 Jun '10

28 Jun '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Fix table_cache negative scalability CREATION DATE..: Fri, 18 Dec 2009, 16:31 SUPERVISOR.....: Bothorsen IMPLEMENTOR....: COPIES TO......: CATEGORY.......: Server-Sprint TASK ID........: 72 (http://askmonty.org/worklog/?tid=72) VERSION........: WorkLog-3.4 STATUS.........: Un-Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: DESCRIPTION: Fix the problem described in this blog entry: http://www.mysqlperformanceblog.com/2009/11/16/table_cache-negative-scalabi… I attempted to paste the contents of the blog here, but worklog didn't accept the text. You have to read the blog entry. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v4.0.0)

1 0

[Maria-developers] WL#73 Deleted (by Serg): Fix table_cache negative scalability
by worklog-noreply＠askmonty.org 28 Jun '10

28 Jun '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Fix table_cache negative scalability CREATION DATE..: Fri, 18 Dec 2009, 16:31 SUPERVISOR.....: Bothorsen IMPLEMENTOR....: COPIES TO......: CATEGORY.......: Server-Sprint TASK ID........: 73 (http://askmonty.org/worklog/?tid=73) VERSION........: WorkLog-3.4 STATUS.........: Un-Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: DESCRIPTION: Fix the problem described in this blog entry: http://www.mysqlperformanceblog.com/2009/11/16/table_cache-negative-scalabi… You can read the blog, or the text below. --- quoted text --- November 16, 2009 table_cache negative scalability Posted by peter | Vote on Planet MySQL Couple of months ago there was a post by FreshBooks on getting great performance improvements by lowering table_cache variable. So I decided to investigate what is really happening here. The common sense approach to tuning caches is to get them as large as you can if you have enough resources (such as memory). With MySQL common sense however does not always works weve seen performance issues with large query_cache_size also sort_buffer_size and read_buffer_size may not give you better performance if you increase them. I found this also applies to some other buffers. Even though having previous experience of surprised behavior I did not expect such a table_cache issue the LRU for cache management is classics and there are scalable algorithms to deal with it. I would expect Monty to implement one of them. To do the test I have created 100.000 empty tables containing single integer column and no indexes and when ran SELECT * FROM tableN in the loop. Each table in such case is accessed only once and on any but first run each access would require table replacement in table cache based on LRU logic. MySQL Sandbox helped me to test this with different servers easily. I did test on CentOS 5.3, Xeon E5405, 16GB RAM and EXT3 file system on the SATA hard drive. MySQL 5.0.85 Created 100.000 tables in around 3min 40 sec which is about 450 tables/sec This indicates the fsync is lying on this test system as default sync_frm option is used. With default table_cache=64 accessing all tables take 12 sec which is almost 8500 tables/sec which is a great speed. We can note significant writes to the disk during this read-only benchmark. Why ? Because for MyISAM tables table header has to be modified each time the table is opened. In this case the performance was so great because all 100.000 tables data (first block of index) was placed close by on disk as well as fully cached which made updates to headers very slow. In the production systems with table headers not in OS cache you often will see significantly low numbers 100 or less. With significantly larger table_cache=16384 (and appropriately adjusted number of open files) the same operation takes 660 seconds which is 151 tables/sec which is around 50 times slower. Wow. This is the slow down. We can see the load becomes very CPU bound in this case and it looks like some of the table_cache algorithms do not scale well. The absolute numbers are also very interesting 151 tables/sec is not that bad if you look at it as an absolute number. So if you tune table cache is normal case and is able to bring down your miss rate (opened_tables) to 10/sec or less by using large table_cache you should do so. However if you have so many tables you still see 100+ misses/sec while your data (at least table headers) is well cached so the cost of table cache miss is not very high, you may be better of with significantly reduced table cache size. The next step for me was to see if the problem was fixed in MySQL 5.1 in this version table_cache was significantly redone and split in table_open_cache and table_definition_cache and I assumed the behavior may be different as well. MySQL 5.1.40 I started testing with default table_open_cache=64 and table_definition_cache=256 the read took about 12 seconds very close to MySQL 5.0.85. As I increased table_definition_cache to 16384 result remained the same so this variable is not causing the bottleneck. However increasing table_open_cache to 16384 causes scan to take about 780 sec which is a bit worse than MySQL 5.0.85. So the problem is not fixed in MySQL 5.1, lets see how MySQL 5.4 behaves. MySQL 5.4.2 MySQL 5.4.2 has higher default table_open_cache so I took it down to 64 so we can compare apples to apples. It performs same as MySQL 5.0 and MySQL 5.1 with small table cache. With table_open_cache increased to 16384 the test took 750 seconds so the problem exists in MySQL 5.4 as well. So the problem is real and it is not fixed even in Performance focused MySQL 5.4. As we can see large table_cache (or table_open_cache_ values indeed can cause significant performance problems. Interesting enough Innodb has a very similar task of managing its own cache of file descriptors (set by innodb_open_files) As the time allows I should test if Heikki knows how to implement LRU properly so it does not have problem with large number. Well see. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v4.0.0)

1 0

[Maria-developers] WL#78 Deleted (by Serg): INSTALL PLUGIN *
by worklog-noreply＠askmonty.org 28 Jun '10

28 Jun '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: INSTALL PLUGIN * CREATION DATE..: Tue, 09 Feb 2010, 18:10 SUPERVISOR.....: Sergei IMPLEMENTOR....: COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 78 (http://askmonty.org/worklog/?tid=78) VERSION........: WorkLog-3.4 STATUS.........: Cancelled PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 10 (hours remain) ORIG. ESTIMATE.: 10 PROGRESS NOTES: -=-=(Serg - Mon, 08 Mar 2010, 20:13)=-=- High Level Description modified. --- /tmp/wklog.78.old.32207 2010-03-08 20:13:06.000000000 +0000 +++ /tmp/wklog.78.new.32207 2010-03-08 20:13:06.000000000 +0000 @@ -6,3 +6,5 @@ INSTALL PLUGIN * SONAME xxx would be a more convenient way to install everything at once. + +cancelled, as a duplicate of mwl:77 -=-=(Serg - Mon, 08 Mar 2010, 20:12)=-=- Status updated. --- /tmp/wklog.78.old.32035 2010-03-08 20:12:39.000000000 +0000 +++ /tmp/wklog.78.new.32035 2010-03-08 20:12:39.000000000 +0000 @@ -1 +1 @@ -Un-Assigned +Cancelled DESCRIPTION: InnoDB, XtraDB, PBXT (at least) come with a storage engine plugin and many information_schema plugins in one .so file. Currently one needs to install them all one by one. INSTALL PLUGIN * SONAME xxx would be a more convenient way to install everything at once. cancelled, as a duplicate of mwl:77 ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v4.0.0)

1 0

[Maria-developers] WL#78 Deleted (by Serg): INSTALL PLUGIN *
by worklog-noreply＠askmonty.org 28 Jun '10

28 Jun '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: INSTALL PLUGIN * CREATION DATE..: Tue, 09 Feb 2010, 18:10 SUPERVISOR.....: Sergei IMPLEMENTOR....: COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 78 (http://askmonty.org/worklog/?tid=78) VERSION........: WorkLog-3.4 STATUS.........: Cancelled PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 10 (hours remain) ORIG. ESTIMATE.: 10 PROGRESS NOTES: -=-=(Serg - Mon, 08 Mar 2010, 20:13)=-=- High Level Description modified. --- /tmp/wklog.78.old.32207 2010-03-08 20:13:06.000000000 +0000 +++ /tmp/wklog.78.new.32207 2010-03-08 20:13:06.000000000 +0000 @@ -6,3 +6,5 @@ INSTALL PLUGIN * SONAME xxx would be a more convenient way to install everything at once. + +cancelled, as a duplicate of mwl:77 -=-=(Serg - Mon, 08 Mar 2010, 20:12)=-=- Status updated. --- /tmp/wklog.78.old.32035 2010-03-08 20:12:39.000000000 +0000 +++ /tmp/wklog.78.new.32035 2010-03-08 20:12:39.000000000 +0000 @@ -1 +1 @@ -Un-Assigned +Cancelled DESCRIPTION: InnoDB, XtraDB, PBXT (at least) come with a storage engine plugin and many information_schema plugins in one .so file. Currently one needs to install them all one by one. INSTALL PLUGIN * SONAME xxx would be a more convenient way to install everything at once. cancelled, as a duplicate of mwl:77 ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v4.0.0)

1 0

[Maria-developers] WL#75 Deleted (by Serg): Extend build to create a shared libmysqld.so library
by worklog-noreply＠askmonty.org 28 Jun '10

28 Jun '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Extend build to create a shared libmysqld.so library CREATION DATE..: Fri, 22 Jan 2010, 09:39 SUPERVISOR.....: Knielsen IMPLEMENTOR....: COPIES TO......: CATEGORY.......: Server-RawIdeaBin TASK ID........: 75 (http://askmonty.org/worklog/?tid=75) VERSION........: Server-9.x STATUS.........: Un-Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: DESCRIPTION: Currently, the embedded library libmysqld is only built as a static library, libmysqld.a. However, there is also a need for a shared embedded library, libmysqld.so. A good example is Amarok, which is now using MySQL (and by default an embedded libmysqld) for storing the user's music collection. Amarok is designed to load a number of modules as .so plugins, and the code using libmysqld is one such plugin. Any code that is to be loaded as a shared object on Linux must be built with -fPIC (position independent code). Any library used must thus also be built with -fPIC and preferably itself be linked statically. Amarok is a widely used application (end-user desktop). It could be a good leverage for making libmysqld more popular eg. in distros. However, currently the distros need to resort to various hacks to make things work with libmysqld and Amarok. Some links: Fedora patches .spec to add -fPIC in the build, extract all objects from libmysqld.a, and re-link them as libmysqld.so. http://cvs.fedoraproject.org/viewvc/devel/mysql/mysql.spec?r1=1.108&r2=1.109 Gentoo apparently have a patch for MySQL to build libmysqld.so: http://bugs.gentoo.org/attachment.cgi?id=188057 Debian seems to suggest building both libmysqld.a and libmysqld_pic.a, the latter to be used for shared objects linking with embedded server: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=508406#52 It would be nice if the main build system was able to build properly libmysqld.so, so different distros would not need to resort to different hacks to get things working. The normal way to build both .a and .so as far as I know is to use libtool; it will then build each object twice (with and without -fPIC) if needed. This one might want to make optional to reduce build times. It would also need to be done for each storage engine. An alternative would be to build everything with -fPIC. This should probably be optional also, as on some architectures (ELF x86 32-bit), there is some speed penalty for -fPIC. The libmysqld.so would then be made only in this case. Distro packages could do a separate build with -fPIC to make libmysqld.so (to not get -fPIC into the main server code). There is a MySQL bug for this: http://bugs.mysql.com/bug.php?id=39288 ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v4.0.0)

1 0

[Maria-developers] WL#84 Deleted (by Serg): Unused
by worklog-noreply＠askmonty.org 28 Jun '10

28 Jun '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Unused CREATION DATE..: Sun, 14 Feb 2010, 00:09 SUPERVISOR.....: Monty IMPLEMENTOR....: COPIES TO......: Igor, Monty, Sergei CATEGORY.......: Server-BackLog TASK ID........: 84 (http://askmonty.org/worklog/?tid=84) VERSION........: Server-9.x STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: -=-=(Igor - Tue, 16 Mar 2010, 19:28)=-=- Title modified. --- /tmp/wklog.84.old.22271 2010-03-16 19:28:50.000000000 +0000 +++ /tmp/wklog.84.new.22271 2010-03-16 19:28:50.000000000 +0000 @@ -1 +1 @@ -Partitioned Key Cache for MyISAM +Unused -=-=(Igor - Tue, 16 Mar 2010, 19:28)=-=- Version updated. --- /tmp/wklog.84.old.22271 2010-03-16 19:28:50.000000000 +0000 +++ /tmp/wklog.84.new.22271 2010-03-16 19:28:50.000000000 +0000 @@ -1 +1 @@ -Benchmarks-3.0 +Server-9.x -=-=(Igor - Tue, 16 Mar 2010, 19:28)=-=- High Level Description modified. --- /tmp/wklog.84.old.22253 2010-03-16 19:28:09.000000000 +0000 +++ /tmp/wklog.84.new.22253 2010-03-16 19:28:09.000000000 +0000 @@ -1,18 +1 @@ -A partitioned key cache is a collection of structures for regular MyiSAM key -caches called key cache partitions. Any page from a file can be placed into a -buffer of only one partition. The number of the partition is calculated from the -file number and the position of the page in the file, and it's always the same -for the page. The function that maps pages into partitions takes care of even -distribution of pages among partitions. -Partition key cache mitigate one of the major problem of simple key cache: -thread contention for key cache lock (mutex). Every call of a key cache -interface function must acquire this lock. So threads compete for this lock even -in the case when they have acquired shared locks for the file and pages they -want read from are in the key cache buffers. When working with a partitioned key -cache any key cache interface function that needs only one page has to acquire -the key cache lock only for the partition the page is ascribed to. This makes -the chances for threads not compete for the same key cache lock better. - -The idea and the original of the partitioned key cache was provided by one of -our external contributers. DESCRIPTION: ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v4.0.0)

1 0

[Maria-developers] WL#84 Deleted (by Serg): Unused
by worklog-noreply＠askmonty.org 28 Jun '10

28 Jun '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Unused CREATION DATE..: Sun, 14 Feb 2010, 00:09 SUPERVISOR.....: Monty IMPLEMENTOR....: COPIES TO......: Igor, Monty, Sergei CATEGORY.......: Server-BackLog TASK ID........: 84 (http://askmonty.org/worklog/?tid=84) VERSION........: Server-9.x STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: -=-=(Igor - Tue, 16 Mar 2010, 19:28)=-=- Title modified. --- /tmp/wklog.84.old.22271 2010-03-16 19:28:50.000000000 +0000 +++ /tmp/wklog.84.new.22271 2010-03-16 19:28:50.000000000 +0000 @@ -1 +1 @@ -Partitioned Key Cache for MyISAM +Unused -=-=(Igor - Tue, 16 Mar 2010, 19:28)=-=- Version updated. --- /tmp/wklog.84.old.22271 2010-03-16 19:28:50.000000000 +0000 +++ /tmp/wklog.84.new.22271 2010-03-16 19:28:50.000000000 +0000 @@ -1 +1 @@ -Benchmarks-3.0 +Server-9.x -=-=(Igor - Tue, 16 Mar 2010, 19:28)=-=- High Level Description modified. --- /tmp/wklog.84.old.22253 2010-03-16 19:28:09.000000000 +0000 +++ /tmp/wklog.84.new.22253 2010-03-16 19:28:09.000000000 +0000 @@ -1,18 +1 @@ -A partitioned key cache is a collection of structures for regular MyiSAM key -caches called key cache partitions. Any page from a file can be placed into a -buffer of only one partition. The number of the partition is calculated from the -file number and the position of the page in the file, and it's always the same -for the page. The function that maps pages into partitions takes care of even -distribution of pages among partitions. -Partition key cache mitigate one of the major problem of simple key cache: -thread contention for key cache lock (mutex). Every call of a key cache -interface function must acquire this lock. So threads compete for this lock even -in the case when they have acquired shared locks for the file and pages they -want read from are in the key cache buffers. When working with a partitioned key -cache any key cache interface function that needs only one page has to acquire -the key cache lock only for the partition the page is ascribed to. This makes -the chances for threads not compete for the same key cache lock better. - -The idea and the original of the partitioned key cache was provided by one of -our external contributers. DESCRIPTION: ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v4.0.0)

1 0

[Maria-developers] WL#84 Deleted (by Serg): Unused
by worklog-noreply＠askmonty.org 28 Jun '10

28 Jun '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Unused CREATION DATE..: Sun, 14 Feb 2010, 00:09 SUPERVISOR.....: Monty IMPLEMENTOR....: COPIES TO......: Igor, Monty, Sergei CATEGORY.......: Server-BackLog TASK ID........: 84 (http://askmonty.org/worklog/?tid=84) VERSION........: Server-9.x STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: (hours remain) ORIG. ESTIMATE.: 0 PROGRESS NOTES: -=-=(Igor - Tue, 16 Mar 2010, 19:28)=-=- Title modified. --- /tmp/wklog.84.old.22271 2010-03-16 19:28:50.000000000 +0000 +++ /tmp/wklog.84.new.22271 2010-03-16 19:28:50.000000000 +0000 @@ -1 +1 @@ -Partitioned Key Cache for MyISAM +Unused -=-=(Igor - Tue, 16 Mar 2010, 19:28)=-=- Version updated. --- /tmp/wklog.84.old.22271 2010-03-16 19:28:50.000000000 +0000 +++ /tmp/wklog.84.new.22271 2010-03-16 19:28:50.000000000 +0000 @@ -1 +1 @@ -Benchmarks-3.0 +Server-9.x -=-=(Igor - Tue, 16 Mar 2010, 19:28)=-=- High Level Description modified. --- /tmp/wklog.84.old.22253 2010-03-16 19:28:09.000000000 +0000 +++ /tmp/wklog.84.new.22253 2010-03-16 19:28:09.000000000 +0000 @@ -1,18 +1 @@ -A partitioned key cache is a collection of structures for regular MyiSAM key -caches called key cache partitions. Any page from a file can be placed into a -buffer of only one partition. The number of the partition is calculated from the -file number and the position of the page in the file, and it's always the same -for the page. The function that maps pages into partitions takes care of even -distribution of pages among partitions. -Partition key cache mitigate one of the major problem of simple key cache: -thread contention for key cache lock (mutex). Every call of a key cache -interface function must acquire this lock. So threads compete for this lock even -in the case when they have acquired shared locks for the file and pages they -want read from are in the key cache buffers. When working with a partitioned key -cache any key cache interface function that needs only one page has to acquire -the key cache lock only for the partition the page is ascribed to. This makes -the chances for threads not compete for the same key cache lock better. - -The idea and the original of the partitioned key cache was provided by one of -our external contributers. DESCRIPTION: ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v4.0.0)

1 0

[Maria-developers] WL#86 Deleted (by Serg): Unused
by worklog-noreply＠askmonty.org 28 Jun '10

28 Jun '10

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Unused CREATION DATE..: Sun, 14 Feb 2010, 00:17 SUPERVISOR.....: Monty IMPLEMENTOR....: Igor COPIES TO......: Igor, Monty, Sergei CATEGORY.......: Server-Sprint TASK ID........: 86 (http://askmonty.org/worklog/?tid=86) VERSION........: Server-5.2 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 0 ESTIMATE.......: 80 (hours remain) ORIG. ESTIMATE.: 80 PROGRESS NOTES: -=-=(Igor - Tue, 16 Mar 2010, 19:30)=-=- Title modified. --- /tmp/wklog.86.old.22309 2010-03-16 19:30:04.000000000 +0000 +++ /tmp/wklog.86.new.22309 2010-03-16 19:30:04.000000000 +0000 @@ -1 +1 @@ -Partitioned Key Cache for MyISAM +Unused -=-=(Igor - Tue, 16 Mar 2010, 19:29)=-=- High Level Description modified. --- /tmp/wklog.86.old.22292 2010-03-16 19:29:37.000000000 +0000 +++ /tmp/wklog.86.new.22292 2010-03-16 19:29:37.000000000 +0000 @@ -1,19 +1 @@ -A partitioned key cache is a collection of structures for regular MyiSAM key -caches called key cache partitions. Any page from a file can be placed into a -buffer of only one partition. The number of the partition is calculated from the -file number and the position of the page in the file, and it's always the same -for the page. The function that maps pages into partitions takes care of even -distribution of pages among partitions. -Partition key cache mitigate one of the major problem of simple key cache: -thread contention for key cache lock (mutex). Every call of a key cache -interface function must acquire this lock. So threads compete for this lock even -in the case when they have acquired shared locks for the file and pages they -want read from are in the key cache buffers. When working with a partitioned key -cache any key cache interface function that needs only one page has to acquire -the key cache lock only for the partition the page is ascribed to. This makes -the chances for threads not compete for the same key cache lock better. - -The idea and the original of the partitioned key cache was provided by one of -our external contributers (see the attached file segmented_keycache_v2.diff with -the original patch from the contributor). -=-=(Igor - Sun, 14 Feb 2010, 00:19)=-=- Privacy level updated. --- /tmp/wklog.86.old.10092 2010-02-13 22:19:03.000000000 +0000 +++ /tmp/wklog.86.new.10092 2010-02-13 22:19:03.000000000 +0000 @@ -1 +1 @@ -y +n -=-=(Igor - Sun, 14 Feb 2010, 00:19)=-=- Category updated. --- /tmp/wklog.86.old.10092 2010-02-13 22:19:03.000000000 +0000 +++ /tmp/wklog.86.new.10092 2010-02-13 22:19:03.000000000 +0000 @@ -1 +1 @@ -Server-BackLog +Server-Sprint -=-=(Igor - Sun, 14 Feb 2010, 00:18)=-=- Version updated. --- /tmp/wklog.86.old.10044 2010-02-14 00:18:31.000000000 +0200 +++ /tmp/wklog.86.new.10044 2010-02-14 00:18:31.000000000 +0200 @@ -1 +1 @@ -Benchmarks-3.0 +Server-5.2 DESCRIPTION: ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v4.0.0)

1 0