# slow performance **URL:** \ **Category:** Percona XtraDB Cluster 8.x\ **Created:** [May 7, 2020, 3:12am UTC](https://forums.percona.com/t/slow-performance/7603 "2020-05-07T03:12:44Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![Ronny\_Gorner](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/ronny_gorner/32/3099_2.png) [@Ronny\_Gorner](https://forums.percona.com/u/Ronny_Gorner)\ **Post date:** [May 7, 2020, 3:12am UTC](https://forums.percona.com/t/slow-performance/7603/1 "2020-05-07T03:12:44Z")
Hi, played around with nearby same settings, on 5.7 and 8.0 xtradb cluster, however, speed in inserts is very very very poor ☹ . Can anyone confirm this ?We use GA first release, CEPH on 2 x 10 GBit network as underlying fs layer, and comparing with 5.7 cluster, speed is now on 25% only than before. We changed collation however, but a 75% breakdown can not be true at all ?BestRonny ---
**Author:** ![Ronny\_Gorner](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/ronny_gorner/32/3099_2.png) [@Ronny\_Gorner](https://forums.percona.com/u/Ronny_Gorner)\ **Post date:** [May 7, 2020, 6:00am UTC](https://forums.percona.com/t/slow-performance/7603/2 "2020-05-07T06:00:41Z")
sysbench oltp\_read\_only --mysql-ssl=off --report-interval=1 --time=300 --threads=4 --tables=10 --table-size=10000000 --mysql-user=xxx --mysql-password=xxx runresult: thds: 4 tps: 3454.67 qps: 55309.68sysbench oltp\_write\_only --mysql-ssl=off --report-interval=1 --time=300 --threads=4 --tables=10 --table-size=10000000 --mysql-user=xxx --mysql-password=xxx runresult: thds: 4 tps: 48.98 qps: 285.86Of course I ran those tests until the end, read is at 55000 qps ~~ and write, around 200 - 300 qps which is poor, poor, poor ☹ and I am out of ideas, I modified multiple variables (including slaves, wsrep, and many more of those innodb parameters) without a luck at all.Network below is a 2 x 10 Gigabit infrastructure on Juniper Networks equipment, and with 5.7 versions, it runs ok. ---
**Author:** ![Marco.Tusa](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/marco.tusa/32/1323_2.png) [@Marco.Tusa](https://forums.percona.com/u/Marco.Tusa)\ **Post date:** [May 7, 2020, 6:41am UTC](https://forums.percona.com/t/slow-performance/7603/3 "2020-05-07T06:41:06Z")
@“Ronny Görner”  can you please share the noede layout you are using and the my.cnf file?  ---
**Author:** ![Ronny\_Gorner](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/ronny_gorner/32/3099_2.png) [@Ronny\_Gorner](https://forums.percona.com/u/Ronny_Gorner)\ **Post date:** [May 7, 2020, 7:38am UTC](https://forums.percona.com/t/slow-performance/7603/4 "2020-05-07T07:38:00Z")
@Marco\_the\_grinch1 I have 3 nodes, all of them are connected over a redundant Juniper network switching infrastructure,on all the config is like this, of course, ips, node-names and so on differ on each of the 3 :😄 # Template my.cnf for PXC # Edit to your requirements. [client] socket=/var/run/mysqld/mysqld.sock default-character-set=utf8mb4 [mysqld] server-id=3 datadir=/var/lib/mysql socket=/var/run/mysqld/mysqld.sock log-error=/var/log/mysql/error.log pid-file=/var/run/mysqld/mysqld.pid # parameter pxc-encrypt-cluster-traffic=OFF sql\_mode=“NO\_ENGINE\_SUBSTITUTION” open\_files\_limit=80000 table\_open\_cache=30000 group\_concat\_max\_len=4096 max\_connections=800 max\_sp\_recursion\_depth=255 join\_buffer\_size=8M read\_buffer\_size=8M sort\_buffer\_size=8M myisam\_sort\_buffer\_size=256M thread\_cache\_size=64 thread\_stack=32M tmp\_table\_size=10G max\_heap\_table\_size=10G table\_definition\_cache=16384 max\_allowed\_packet=256M read\_rnd\_buffer\_size=8M preload\_buffer\_size=16M net\_buffer\_length=16M query\_prealloc\_size=16M table\_definition\_cache=30000 innodb\_read\_io\_threads = 64 innodb\_write\_io\_threads = 64 innodb\_io\_capacity = 5000 innodb\_io\_capacity\_max = 10000 innodb\_thread\_concurrency=0 innodb\_buffer\_pool\_size=64G innodb\_log\_buffer\_size=512M innodb\_log\_file\_size=1G innodb\_concurrency\_tickets=8192 innodb\_open\_files=80000 init\_connect=‘SET COLLATION\_CONNECTION=utf8mb4\_0900\_ai\_ci’ init\_connect=‘SET NAMES utf8mb4 COLLATE utf8mb4\_0900\_ai\_ci’ collation\_server=utf8mb4\_0900\_ai\_ci character-set-server=utf8mb4 skip-character-set-client-handshake # Binary log expiration period is 604800 seconds, which equals 7 days binlog\_expire\_logs\_seconds=604800 # Path to Galera library wsrep\_provider=/usr/lib/galera4/libgalera\_smm.so # Cluster connection URL contains IPs of nodes #If no IP is found, this implies that a new cluster needs to be created, #in order to do that you need to bootstrap this node wsrep\_cluster\_address=gcomm://1.2.3.4,1.2.3.5,1.2.3.6 # In order for Galera to work correctly binlog format should be ROW binlog\_format=ROW # This changes how InnoDB autoincrement locks are managed and is a requirement for Galera innodb\_autoinc\_lock\_mode=2 # Node IP address wsrep\_node\_address=1.2.3.4 # Cluster name wsrep\_cluster\_name=my-cluster #If wsrep\_node\_name is not specified,  then system hostname will be used wsrep\_node\_name=my-cluster-node-n #pxc\_strict\_mode allowed values: DISABLED,PERMISSIVE,ENFORCING,MASTER pxc\_strict\_mode=ENFORCING # SST method wsrep\_sst\_method=xtrabackup-v2 # limit to 1/2 of cores wsrep\_slave\_threads=16 wsrep\_log\_conflicts # size of the cache for the replication - default is very low wsrep\_provider\_options=“gcache.size=4G; gcache.page\_size=2G;” ---
**Author:** ![Ronny\_Gorner](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/ronny_gorner/32/3099_2.png) [@Ronny\_Gorner](https://forums.percona.com/u/Ronny_Gorner)\ **Post date:** [May 7, 2020, 9:40am UTC](https://forums.percona.com/t/slow-performance/7603/5 "2020-05-07T09:40:52Z")
Switching FS from btrfs to ext4, now we have:thds: 4 tps: 253.59 qps: 1541.48 ---
**Author:** ![Ronny\_Gorner](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/ronny_gorner/32/3099_2.png) [@Ronny\_Gorner](https://forums.percona.com/u/Ronny_Gorner)\ **Post date:** [May 7, 2020, 11:30am UTC](https://forums.percona.com/t/slow-performance/7603/6 "2020-05-07T11:30:34Z")
Running on 16 wsrep\_slave\_threads, and having 32 threads writting on sysbench, we have around 4000 qps. However, more than 16 threads does not seem to make any sense because of slave-limit of 16. Docs are confusing, on the hand, it should be half of available cpu cores, on the other hand, nearby the max amount of connections. So?