 |
|
|
|
Next: stripping off info after last common
|
| Author |
Message |
External
Since: May 26, 2008
Posts: 15
|
(Msg. 1) Posted: Thu Nov 06, 2008 12:25 pm
Post subject: High Load Average
Archived from groups: comp>databases>ingres (more info?)
|
|
|
Hi Guys,
I have a box with 16 CPU´s, 16Gb of RAM and a EVA3000 with RAID5. My
load average is high, but the strange is that I have CPU´s idle.
Look down a example, I have one DBMS using 201% of CPU (top), but I have
other with 100% Idle ! At the same time, I have 5,6,10 proccess waiting
for CPU (vmstat)
How can I balance my load to the other CPU´s ? Adding more DBMS´s Servers ?
Tkz All !
model name : Intel(R) Xeon(R) CPU X7350 @ 2.93GHz
free
total used free shared buffers cached
Mem: 16409896 16341548 68348 0 141340 13754024
-/+ buffers/cache: 2446184 13963712
Swap: 8388600 260 8388340
top - 14:28:24 up 2 days, 6:57, 2 users, load average: 3.78, 4.97, 5.14
Tasks: 195 total, 1 running, 184 sleeping, 0 stopped, 10 zombie
Cpu0 : 4.7% us, 2.7% sy, 0.0% ni, 92.3% id, 0.0% wa, 0.0% hi, 0.3% si
Cpu1 : 61.6% us, 34.0% sy, 0.0% ni, 4.4% id, 0.0% wa, 0.0% hi, 0.0% si
Cpu2 : 0.0% us, 0.0% sy, 0.0% ni, 100.0% id, 0.0% wa, 0.0% hi,
0.0% si
Cpu3 : 2.0% us, 2.7% sy, 0.0% ni, 95.3% id, 0.0% wa, 0.0% hi, 0.0% si
Cpu4 : 1.7% us, 0.0% sy, 0.0% ni, 98.0% id, 0.3% wa, 0.0% hi, 0.0% si
Cpu5 : 0.0% us, 0.0% sy, 0.0% ni, 100.0% id, 0.0% wa, 0.0% hi,
0.0% si
Cpu6 : 1.3% us, 0.3% sy, 0.0% ni, 98.3% id, 0.0% wa, 0.0% hi, 0.0% si
Cpu7 : 2.0% us, 0.7% sy, 0.0% ni, 97.0% id, 0.3% wa, 0.0% hi, 0.0% si
Cpu8 : 0.3% us, 0.0% sy, 0.0% ni, 99.7% id, 0.0% wa, 0.0% hi, 0.0% si
Cpu9 : 8.0% us, 0.7% sy, 0.0% ni, 91.0% id, 0.0% wa, 0.0% hi, 0.3% si
Cpu10 : 0.3% us, 0.0% sy, 0.0% ni, 99.7% id, 0.0% wa, 0.0% hi, 0.0% si
Cpu11 : 0.0% us, 0.0% sy, 0.0% ni, 100.0% id, 0.0% wa, 0.0% hi,
0.0% si
Cpu12 : 6.0% us, 0.7% sy, 0.0% ni, 71.1% id, 22.3% wa, 0.0% hi, 0.0% si
Cpu13 : 0.3% us, 0.0% sy, 0.0% ni, 99.7% id, 0.0% wa, 0.0% hi, 0.0% si
Cpu14 : 0.3% us, 0.3% sy, 0.0% ni, 99.3% id, 0.0% wa, 0.0% hi, 0.0% si
Cpu15 : 63.5% us, 32.8% sy, 0.0% ni, 3.8% id, 0.0% wa, 0.0% hi, 0.0% si
Mem: 16409896k total, 16336652k used, 73244k free, 136828k buffers
Swap: 8388600k total, 260k used, 8388340k free, 13750280k cached
PID USER PR NI %CPU TIME+ %MEM VIRT RES SHR S COMMAND
10544 ingres 16 0 201 1202:49 11.8 2710m 1.9g 1.6g S
/strip3/ingres/bin/iidbms dbms (default) II
10612 ingres 16 0 12 1149:39 11.7 2695m 1.8g 1.5g S
/strip3/ingres/bin/iidbms dbms (default) II
10000 ingres 16 0 5 915:17.70 11.8 2708m 1.9g 1.6g S
/strip3/ingres/bin/iidbms dbms (default) II
10477 ingres 16 0 2 765:48.42 11.3 2703m 1.8g 1.5g S
/strip3/ingres/bin/iidbms dbms (default) II
10739 ingres 16 0 1 16:47.06 0.0 13784 5796 904 S
/strip3/ingres/bin/iigcc II gcc
10907 ingres 16 0 1 17:24.20 0.0 13768 5772 904 S
/strip3/ingres/bin/iigcc II gcc
10873 ingres 16 0 0 15:44.65 0.0 13832 5840 904 S
/strip3/ingres/bin/iigcc II gcc
[root@sel-dip-01 ~]# vmstat 1
procs -----------memory---------- ---swap-- -----io---- --system--
----cpu----
r b swpd free buff cache si so bi bo in cs us
sy id wa
2 0 260 73428 136908 13750716 0 0 119 89 7 10 7
2 91 1
4 0 260 73420 136912 13750712 0 0 20 368 1805 40342 14
5 81 0
2 1 260 64588 136956 13757376 0 0 28 46940 7693 58546 13
7 67 13
2 1 260 46852 136920 13773924 0 0 240 39436 2842 44517 11
6 81 2
2 0 260 48572 136928 13773916 0 0 68 172 1681 41182 9
4 84 2
1 0 260 48380 136936 13773908 0 0 260 96 1600 41001 10
4 86 0
2 0 260 48516 136944 13773900 0 0 32 8984 1666 38565 9
4 86 1
3 0 260 48452 136944 13773900 0 0 20 32 1294 38522 10
4 86 0
3 0 260 48460 136952 13774408 0 0 12 336 1686 39274 9
4 86 0
3 0 260 48508 136960 13774400 0 0 4 128 2125 40357 13
5 82 0
2 0 260 48396 136960 13774400 0 0 60 84 1975 40933 14
5 81 0
3 0 260 48276 136960 13774916 0 0 72 64 1541 38551 14
5 81 0
2 0 260 48084 136968 13774908 0 0 72 128 1779 39052 12
4 84 0
5 0 260 47964 136984 13774892 0 0 32 384 1789 39067 11
5 83 0
6 0 260 47740 136984 13775924 0 0 352 116 2613 52954 16
9 74 1
5 0 260 47484 136992 13776432 0 0 24 348 2275 60933 15
10 75 0
5 0 260 47108 136992 13776432 0 0 20 64 1702 61104 18
9 73 0
--
Regards,
Thiago Zerbinato
Oracle/Ingres DBA
Ribeirao Preto/SP
Brazil |
|
| Back to top |
|
 | |
External
Since: Jul 21, 2004
Posts: 121
|
(Msg. 2) Posted: Thu Nov 06, 2008 1:14 pm
Post subject: Re: [Info-Ingres] High Load Average [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)
|
|
|
On Nov 6, 2008, at 12:41 PM, thiagomz wrote:
> Oscar Carlés escreveu:
>> Hi Thiago!
>> What is your Ingres and OS version?
>> Do you use os threads or internal ingres slaves?
>>
>> Oscar
>>
>
> [ingres@sel-dip-01 ingres]$ cat version.rel
> II 2.6/0604 (a64.lnx/00)
> 12292
> ------------------------------------------------------------
> II 2.6/0604 (a64.lnx/00)
> SP5
>
Which gives you your answer.
Ingres 2.6 on Linux uses "simulate process shared", which is a clumsy
and inefficient way of working around some cross-process mutexing and
signalling limitations that older linux kernels had. Ingres 2006
will run better, since it is built for NPTL kernels (ie linux 2.6+)
and doesn't have the simulate process shared stuff.
The best solution is to upgrade, but as far as 2.6 is concerned,
you could possibly try running more DBMS server instances to
see if you can get the other CPU's into the act.
Karl |
|
| Back to top |
|
| |
External
Since: May 03, 2005
Posts: 13
|
(Msg. 3) Posted: Thu Nov 06, 2008 1:26 pm
Post subject: Re: [Info-Ingres] High Load Average [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)
|
|
|
Hi Thiago!
What is your Ingres and OS version?
Do you use os threads or internal ingres slaves?
Oscar
thiagomz wrote:
> Hi Guys,
>
> I have a box with 16 CPU´s, 16Gb of RAM and a EVA3000 with RAID5. My
> load average is high, but the strange is that I have CPU´s idle.
>
> Look down a example, I have one DBMS using 201% of CPU (top), but I have
> other with 100% Idle ! At the same time, I have 5,6,10 proccess waiting
> for CPU (vmstat)
>
> How can I balance my load to the other CPU´s ? Adding more DBMS´s Servers ?
>
>
> Tkz All !
>
>
> model name : Intel(R) Xeon(R) CPU X7350 @ 2.93GHz
>
> free
> total used free shared buffers cached
> Mem: 16409896 16341548 68348 0 141340 13754024
> -/+ buffers/cache: 2446184 13963712
> Swap: 8388600 260 8388340
>
>
> top - 14:28:24 up 2 days, 6:57, 2 users, load average: 3.78, 4.97, 5.14
> Tasks: 195 total, 1 running, 184 sleeping, 0 stopped, 10 zombie
> Cpu0 : 4.7% us, 2.7% sy, 0.0% ni, 92.3% id, 0.0% wa, 0.0% hi, 0.3% si
> Cpu1 : 61.6% us, 34.0% sy, 0.0% ni, 4.4% id, 0.0% wa, 0.0% hi, 0.0% si
> Cpu2 : 0.0% us, 0.0% sy, 0.0% ni, 100.0% id, 0.0% wa, 0.0% hi,
> 0.0% si
> Cpu3 : 2.0% us, 2.7% sy, 0.0% ni, 95.3% id, 0.0% wa, 0.0% hi, 0.0% si
> Cpu4 : 1.7% us, 0.0% sy, 0.0% ni, 98.0% id, 0.3% wa, 0.0% hi, 0.0% si
> Cpu5 : 0.0% us, 0.0% sy, 0.0% ni, 100.0% id, 0.0% wa, 0.0% hi,
> 0.0% si
> Cpu6 : 1.3% us, 0.3% sy, 0.0% ni, 98.3% id, 0.0% wa, 0.0% hi, 0.0% si
> Cpu7 : 2.0% us, 0.7% sy, 0.0% ni, 97.0% id, 0.3% wa, 0.0% hi, 0.0% si
> Cpu8 : 0.3% us, 0.0% sy, 0.0% ni, 99.7% id, 0.0% wa, 0.0% hi, 0.0% si
> Cpu9 : 8.0% us, 0.7% sy, 0.0% ni, 91.0% id, 0.0% wa, 0.0% hi, 0.3% si
> Cpu10 : 0.3% us, 0.0% sy, 0.0% ni, 99.7% id, 0.0% wa, 0.0% hi, 0.0% si
> Cpu11 : 0.0% us, 0.0% sy, 0.0% ni, 100.0% id, 0.0% wa, 0.0% hi,
> 0.0% si
> Cpu12 : 6.0% us, 0.7% sy, 0.0% ni, 71.1% id, 22.3% wa, 0.0% hi, 0.0% si
> Cpu13 : 0.3% us, 0.0% sy, 0.0% ni, 99.7% id, 0.0% wa, 0.0% hi, 0.0% si
> Cpu14 : 0.3% us, 0.3% sy, 0.0% ni, 99.3% id, 0.0% wa, 0.0% hi, 0.0% si
> Cpu15 : 63.5% us, 32.8% sy, 0.0% ni, 3.8% id, 0.0% wa, 0.0% hi, 0.0% si
> Mem: 16409896k total, 16336652k used, 73244k free, 136828k buffers
> Swap: 8388600k total, 260k used, 8388340k free, 13750280k cached
>
> PID USER PR NI %CPU TIME+ %MEM VIRT RES SHR S COMMAND
> 10544 ingres 16 0 201 1202:49 11.8 2710m 1.9g 1.6g S
> /strip3/ingres/bin/iidbms dbms (default) II
> 10612 ingres 16 0 12 1149:39 11.7 2695m 1.8g 1.5g S
> /strip3/ingres/bin/iidbms dbms (default) II
> 10000 ingres 16 0 5 915:17.70 11.8 2708m 1.9g 1.6g S
> /strip3/ingres/bin/iidbms dbms (default) II
> 10477 ingres 16 0 2 765:48.42 11.3 2703m 1.8g 1.5g S
> /strip3/ingres/bin/iidbms dbms (default) II
> 10739 ingres 16 0 1 16:47.06 0.0 13784 5796 904 S
> /strip3/ingres/bin/iigcc II gcc
> 10907 ingres 16 0 1 17:24.20 0.0 13768 5772 904 S
> /strip3/ingres/bin/iigcc II gcc
> 10873 ingres 16 0 0 15:44.65 0.0 13832 5840 904 S
> /strip3/ingres/bin/iigcc II gcc
>
>
> [root@sel-dip-01 ~]# vmstat 1
> procs -----------memory---------- ---swap-- -----io---- --system--
> ----cpu----
> r b swpd free buff cache si so bi bo in cs us
> sy id wa
> 2 0 260 73428 136908 13750716 0 0 119 89 7 10 7
> 2 91 1
> 4 0 260 73420 136912 13750712 0 0 20 368 1805 40342 14
> 5 81 0
> 2 1 260 64588 136956 13757376 0 0 28 46940 7693 58546 13
> 7 67 13
> 2 1 260 46852 136920 13773924 0 0 240 39436 2842 44517 11
> 6 81 2
> 2 0 260 48572 136928 13773916 0 0 68 172 1681 41182 9
> 4 84 2
> 1 0 260 48380 136936 13773908 0 0 260 96 1600 41001 10
> 4 86 0
> 2 0 260 48516 136944 13773900 0 0 32 8984 1666 38565 9
> 4 86 1
> 3 0 260 48452 136944 13773900 0 0 20 32 1294 38522 10
> 4 86 0
> 3 0 260 48460 136952 13774408 0 0 12 336 1686 39274 9
> 4 86 0
> 3 0 260 48508 136960 13774400 0 0 4 128 2125 40357 13
> 5 82 0
> 2 0 260 48396 136960 13774400 0 0 60 84 1975 40933 14
> 5 81 0
> 3 0 260 48276 136960 13774916 0 0 72 64 1541 38551 14
> 5 81 0
> 2 0 260 48084 136968 13774908 0 0 72 128 1779 39052 12
> 4 84 0
> 5 0 260 47964 136984 13774892 0 0 32 384 1789 39067 11
> 5 83 0
> 6 0 260 47740 136984 13775924 0 0 352 116 2613 52954 16
> 9 74 1
> 5 0 260 47484 136992 13776432 0 0 24 348 2275 60933 15
> 10 75 0
> 5 0 260 47108 136992 13776432 0 0 20 64 1702 61104 18
> 9 73 0
>
>
>
>
--
Lic. Oscar Carlés Barriocanal
Director de Servicios al Cliente
Íntegra S.R.L.
Telefax: (595 21) 424473 RA
Asunción - Paraguay >> Stay informed about: High Load Average |
|
| Back to top |
|
| |
External
Since: May 26, 2008
Posts: 15
|
(Msg. 4) Posted: Thu Nov 06, 2008 1:26 pm
Post subject: Re: [Info-Ingres] High Load Average [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)
|
|
|
Oscar Carlés escreveu:
> Hi Thiago!
> What is your Ingres and OS version?
> Do you use os threads or internal ingres slaves?
>
> Oscar
>
[ingres@sel-dip-01 ingres]$ cat version.rel
II 2.6/0604 (a64.lnx/00)
12292
------------------------------------------------------------
II 2.6/0604 (a64.lnx/00)
SP5
[ingres@sel-dip-01 ingres]$ cat /etc/redhat-release
Red Hat Enterprise Linux AS release 4 (Nahant Update 6)
[ingres@sel-dip-01 ingres]$ uname -a
Linux sel-dip-01.sev.intra 2.6.9-67.ELlargesmp #1 SMP Wed Nov 7 14:07:22
EST 2007 x86_64 x86_64 x86_64 GNU/Linux
--
Regards,
Thiago Zerbinato
Oracle/Ingres DBA
Ribeirao Preto/SP
Brazil >> Stay informed about: High Load Average |
|
| Back to top |
|
| |
External
Since: May 26, 2008
Posts: 15
|
(Msg. 5) Posted: Thu Nov 06, 2008 2:25 pm
Post subject: Re: [Info-Ingres] High Load Average [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)
|
|
|
Karl & Betty Schendel escreveu:
> Which gives you your answer.
>
> Ingres 2.6 on Linux uses "simulate process shared", which is a clumsy
> and inefficient way of working around some cross-process mutexing and
> signalling limitations that older linux kernels had. Ingres 2006
> will run better, since it is built for NPTL kernels (ie linux 2.6+)
> and doesn't have the simulate process shared stuff.
>
> The best solution is to upgrade, but as far as 2.6 is concerned,
> you could possibly try running more DBMS server instances to
> see if you can get the other CPU's into the act.
>
> Karl
>
>
Karl,
Thank you for the answer.
I´m waiting for the new release 9.2, it has something like a standby
database (on oracle). I need to create a backup site and this feature is
important to me.
Cheers !
--
Regards,
Thiago Zerbinato
Oracle/Ingres DBA
Ribeirao Preto/SP
Brazil >> Stay informed about: High Load Average |
|
| Back to top |
|
| |
External
Since: Oct 01, 2008
Posts: 5
|
(Msg. 6) Posted: Thu Nov 06, 2008 10:05 pm
Post subject: Re: [Info-Ingres] High Load Average [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)
|
|
|
Hi
We ran Ingres 2.6 on Linux until about 2 months ago.
Ingres 2006 with NPTL made a huge difference in our installation. With 2.6, load was always over 10
with CPU activity mostly in the USER state.
With 2006 NPTL, the system handles 3 times the user load and the CPUs barely get warm.
CPU state has shifted from USER to IOWAIT. Now I need faster disks 
Andre
________________________________
From: info-ingres-bounces DeleteThis @kettleriverconsulting.com on behalf of Karl & Betty Schendel
Sent: Thu 06/11/2008 1:14 PM
To: Ingres and related product discussion forum
Subject: Re: [Info-Ingres] High Load Average
On Nov 6, 2008, at 12:41 PM, thiagomz wrote:
> Oscar Carlés escreveu:
>> Hi Thiago!
>> What is your Ingres and OS version?
>> Do you use os threads or internal ingres slaves?
>>
>> Oscar
>>
>
> [ingres@sel-dip-01 ingres]$ cat version.rel
> II 2.6/0604 (a64.lnx/00)
> 12292
> ------------------------------------------------------------
> II 2.6/0604 (a64.lnx/00)
> SP5
>
Which gives you your answer.
Ingres 2.6 on Linux uses "simulate process shared", which is a clumsy
and inefficient way of working around some cross-process mutexing and
signalling limitations that older linux kernels had. Ingres 2006
will run better, since it is built for NPTL kernels (ie linux 2.6+)
and doesn't have the simulate process shared stuff.
The best solution is to upgrade, but as far as 2.6 is concerned,
you could possibly try running more DBMS server instances to
see if you can get the other CPU's into the act.
Karl
_______________________________________________
Info-Ingres mailing list
Info-Ingres DeleteThis @kettleriverconsulting.com
http://www.kettleriverconsulting.com/mailman/listinfo/info-ingres >> Stay informed about: High Load Average |
|
| Back to top |
|
| |
External
Since: Oct 01, 2008
Posts: 5
|
(Msg. 7) Posted: Fri Nov 07, 2008 9:20 am
Post subject: Re: [Info-Ingres] High Load Average [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)
|
|
|
No unload/reload, that would've taken days.
The install does an upgradedb and it worked fine.
Our front-end is OpenRoad on XP. They are still running the 2.6 client, for now.
-----Original Message-----
From: info-ingres-bounces RemoveThis @kettleriverconsulting.com [mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of thiagomz
Sent: November 7, 2008 8:55 AM
To: info-ingres RemoveThis @kettleriverconsulting.com
Subject: Re: [Info-Ingres] High Load Average
Laframboise escreveu:
> Hi
> We ran Ingres 2.6 on Linux until about 2 months ago.
>
Laframboise,
What was your method for the migration ?
unload and load ?
Did you get same error with your procedures, functions, etc... I´m using
4gl and Powerbuilder 10.x accessing ingres 2.6
--
Regards,
Thiago Zerbinato
Oracle/Ingres DBA
Ribeirao Preto/SP
Brazil
_______________________________________________
Info-Ingres mailing list
Info-Ingres RemoveThis @kettleriverconsulting.com
http://www.kettleriverconsulting.com/mailman/listinfo/info-ingres >> Stay informed about: High Load Average |
|
| Back to top |
|
| |
External
Since: May 26, 2008
Posts: 15
|
(Msg. 8) Posted: Fri Nov 07, 2008 9:25 am
Post subject: Re: [Info-Ingres] High Load Average [Login to view extended thread Info.]
Archived from groups: per prev. post (more info?)
|
|
|
Laframboise escreveu:
> Hi
> We ran Ingres 2.6 on Linux until about 2 months ago.
>
Laframboise,
What was your method for the migration ?
unload and load ?
Did you get same error with your procedures, functions, etc... I´m using
4gl and Powerbuilder 10.x accessing ingres 2.6
--
Regards,
Thiago Zerbinato
Oracle/Ingres DBA
Ribeirao Preto/SP
Brazil >> Stay informed about: High Load Average |
|
| Back to top |
|
| |
| Related Topics: | Replicator option for high availability and load balancing - Hi, We want to use replication (with INGRES 2006 on Linux) to configure 2 active servers for high availability and load balancing. The 2 target types are full peer, the replicator servers start automatically with the application. To start the..
Bulk Load Stratagies - Bulk Load Stratagies Ok ok... I'm having a brain fart here. (normal I know...) One of the DBA's I work with is trying to reload about 1.2 mil rows into a new table under VMS Ingres 2.0 and is encountering "force abort" errors. (i.e. the rel...
Clustering, Security, Performance, Load Balance - I think this question has been asked number of times. However, I am looking for some specific information. Perhaps some of you can help close the gap. Or perhaps you can point me towards right direction. Perhaps this group can help me fill in Ingress..
Process hanging on Endselect - We have a process which needs to produce an average figure over the last 16 qualifying weeks. The way it currently works is to select the rows from the two relevant tables ordered by date descending, then in a select loop check if the week qualifies and...
E_DM9063_LK_TABLE_MAXLOCKS - We have a weekly job that takes a COPY.OUT from the "Live" database and then does a COPY.IN to a development database in a separate installation. I've noticed that in one instance, the ERRLOG is getting a number of messages of the general form... |
|
You can post new topics in this forum
You can reply to topics in this forum
You can edit your posts in this forum
You can delete your posts in this forum
You can vote in polls in this forum
|
|
|
|
|
|
|
FAQ
Search
Profile
Private Messages
Log in
