ifx performace: linux vs NT

This is a multi-part message in MIME format. ------=_NextPart_000_001B_01BF5CF2.35C4E380 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Palmer: thanks for your input in this isssue. Let me tell you about the tests: I started with a set of queries against 5 tables: one with 150.000 rows, other with 250.000 rows and the rest pretty small ones. First, forcing the engine to sequential scan the whole table. Then, queries of a single table, with a filter condition on a indexed field, with and w/o the order by clause. And last, a series of joins with filters conditions with and w/o the order by clause. In most of the tests the linux box was slower...twice I must say. For example, the sequential scan took NT 1 min and linux 2. I was going to continue with a series of more elaborated tests, but with this figures I panic and post my onconfig for help. I belive I must be doing something wrong. Anyone with a bright idea? System: Linux RedHat 6.0 / Kernel 2.2.5-15 Informix IDS 7.30 UC7 PII 350 Mhz, single cpu 128 RAM NT 4.0 Informix IDS 7.30 TC 5 same HW Cooked files on both, (unbuffered on NT). Statistics updated. ----- Mensaje original ----- De: richard palmer Para: Diego CC: Informix-list Enviado: Martes, 11 de Enero de 2000 04:38 p.m. Asunto: Re: special thanks and informix performance considerations I took a look at the onconfig you attached. When you say you are disappointed with the performance, what manner of testing did you use ? Without an idea of your metrics, performance tuning is more a shot in the dark as opposed to dealing with identified problem areas and resolving these in an orderly manner. I can state that (last I heard) you could not configure more than one AIO VP on an NT system. This in itself can create a bottleneck problem in terms of I/O. The onconfig I received showed NO entry for AIO VPs, which defaults to one. Configuration parameters that work efficiently on NT may not be what works best on a unix based system like Linux. Try bumping up the NUMAIOVPs to 1 per chunk or 2 per dbspace (opinions vary on this, try both ?) to distribute the I/O. Also, try bumping up CLEANERs to atleast one per chunk with a minimum of two. have a nice day. ------=_NextPart_000_001B_01BF5CF2.35C4E380 Content-Type: application/octet-stream; name="onconfig" Content-Transfer-Encoding: quoted-printable Content-Disposition: attachment; filename="onconfig" #************************************************************************= **=0A= #=0A= # INFORMIX SOFTWARE, INC.=0A= #=0A= # Title: onconfig.std=0A= # Description: Informix Dynamic Server Configuration Parameters=0A= #=0A= #************************************************************************= **=0A= =0A= # Root Dbspace Configuration=0A= =0A= ROOTNAME rootdbs # Root dbspace name=0A= ROOTPATH /info1/rootdbs.000 # Path for device containing root = dbspace=0A= ROOTOFFSET 0 # Offset of root dbspace into device = (Kbytes)=0A= ROOTSIZE 50000 # Size of root dbspace (Kbytes)=0A= =0A=# Disk Mirroring Configuration Parameters=0A= =0A= MIRROR 0 # Mirroring flag (Yes =3D 1, No =3D 0)=0A= MIRRORPATH # Path for device containing mirrored = root=0A= MIRROROFFSET 0 # Offset into mirrored device (Kbytes)=0A= =0A=# Physical Log Configuration=0A= =0A= PHYSDBS rootdbs # Location (dbspace) of physical log=0A= PHYSFILE 1000 # Physical log file size (Kbytes)=0A= =0A=# Logical Log Configuration=0A= =0A= LOGFILES 6 # Number of logical log files=0A= LOGSIZE 500 # Logical log size (Kbytes)=0A= =0A= # Diagnostics=0A= =0A=MSGPATH /opt/informix/online.log # System message log file path=0A= CONSOLE /dev/null # System console message path=0A= ALARMPROGRAM /opt/informix/etc/log_full.sh # Alarm program path=0A= SYSALARMPROGRAM /opt/informix/etc/evidence.sh # System Alarm program path=0A= TBLSPACE_STATS 1 =0A= =0A=# System Archive Tape Device=0A= =0A= TAPEDEV /dev/null # Tape device path =0A= TAPEBLK 16 # Tape block size (Kbytes)=0A= TAPESIZE 10240 # Maximum amount of data to put on tape = (Kbytes)=0A= =0A= # Log Archive Tape Device=0A= =0A= LTAPEDEV /dev/null # Log tape device path=0A= LTAPEBLK 16 # Log tape block size (Kbytes)=0A= LTAPESIZE 10240 # Max amount of data to put on log tape = (Kbytes)=0A= =0A= # Optical=0A= =0A=STAGEBLOB # Informix Dynamic Server/Optical = staging area=0A= =0A= # System Configuration=0A= =0A= SERVERNUM 0 # Unique id corresponding to a Dynamic = Server instance=0A= DBSERVERNAME ol_linuxshm # Name of default database server=0A= DBSERVERALIASES ol_linux # List of alternate dbservernames=0A= NETTYPE ipcshm,1,10,CPU # Configure poll thread(s) for nettype=0A= NETTYPE soctcp,1,10,NET # Configure poll thread(s) for nettype=0A= DEADLOCK_TIMEOUT 60 # Max time to wait of lock in = distributed env.=0A= RESIDENT 0 # Forced residency flag (Yes =3D 1, No = =3D 0)=0A= =0A= MULTIPROCESSOR 0 # 0 for single-processor, 1 for = multi-processor=0A= NUMCPUVPS 1 # Number of user (cpu) vps=0A= SINGLE_CPU_VP 0 # If non-zero, limit number of cpu vps = to one=0A= =0A=#NOAGE 0 # Process aging=0A= #AFF_SPROC 0 # Affinity start processor=0A= #AFF_NPROCS 0 # Affinity number of processors=0A= =0A= # Shared Memory Parameters=0A= =0A= LOCKS 2000 # Maximum number of locks=0A= BUFFERS 10000 # Maximum number of shared buffers=0A= NUMAIOVPS # Number of IO vps=0A= PHYSBUFF 32 # Physical log buffer size (Kbytes)=0A= LOGBUFF 32 # Logical log buffer size (Kbytes)=0A=LOGSMAX 6 # Maximum number of logical log files=0A= CLEANERS 1 # Number of buffer cleaner processes=0A= SHMBASE 0x10000000 # Shared memory base address=0A= SHMVIRTSIZE 8000 # initial virtual shared memory segment = size=0A= SHMADD 8192 # Size of new shared memory segments = (Kbytes)=0A= SHMTOTAL 0 # Total shared memory (Kbytes). = 0=3D>unlimited=0A= CKPTINTVL 300 # Check point interval (in sec)=0A= LRUS 4 # Number of LRU queues=0A= LRU_MAX_DIRTY 60 # LRU percent dirty begin cleaning limit=0A= LRU_MIN_DIRTY 50 # LRU percent dirty end cleaning limit=0A= LTXHWM 50 # Long transaction high water mark = percentage=0A= LTXEHWM 60 # Long transaction high water m

Hi, a few remarks regarding your ONCONFIG: * SINGLE_CPU_VP should be set to 1, since you are not using multiple CPU VPs - this should improve performance for CPU intensive tasks * on NT KAIO is used automatically, therefore there is no need for AIOVP on NT. Linux on the other hand doesn't support raw devices (yet) and there is no KAIO. Therefore, I would recommend to increase the number of AIOVPs to be able to handle multiple IO requests in parallel (depending on the number of disks). Hope this helps, Heiko Diego wrote: > This is a multi-part message in MIME format. > > ------=_NextPart_000_001B_01BF5CF2.35C4E380 > Content-Type: text/plain; > charset="iso-8859-1" > Content-Transfer-Encoding: 7bit > > Palmer: thanks for your input in this isssue. Let me tell you about the > tests: > > I started with a set of queries against 5 tables: one with 150.000 rows, > other with 250.000 rows and the rest pretty small ones. > > First, forcing the engine to sequential scan the whole table. > Then, queries of a single table, with a filter condition on a indexed field, > with and w/o the order by clause. > And last, a series of joins with filters conditions with and w/o the order > by clause. > > In most of the tests the linux box was slower...twice I must say. For > example, the sequential scan took NT 1 min and linux 2. > > I was going to continue with a series of more elaborated tests, but with > this figures I panic and post my onconfig for help. I belive I must be doing > something wrong. Anyone with a bright idea? > > System: > > Linux RedHat 6.0 / Kernel 2.2.5-15 > Informix IDS 7.30 UC7 > PII 350 Mhz, single cpu > 128 RAM > > NT 4.0 > Informix IDS 7.30 TC 5 > same HW > > Cooked files on both, (unbuffered on NT). Statistics updated. > > ----- Mensaje original ----- > De: richard palmer > Para: Diego > CC: Informix-list > Enviado: Martes, 11 de Enero de 2000 04:38 p.m. > Asunto: Re: special thanks and informix performance considerations > > I took a look at the onconfig you attached. When you say you are > disappointed with the performance, what manner of testing did you > use ? Without an idea of your metrics, performance tuning is more a > shot in the dark as opposed to dealing with identified problem areas > and resolving these in an orderly manner. > I can state that (last I heard) you could not configure more than one > AIO VP on an NT system. This in itself can create a bottleneck > problem in terms of I/O. The onconfig I received showed NO entry > for AIO VPs, which defaults to one. > Configuration parameters that work efficiently on NT may not be > what works best on a unix based system like Linux. Try bumping > up the NUMAIOVPs to 1 per chunk or 2 per dbspace (opinions > vary on this, try both ?) to distribute the I/O. Also, try bumping up > CLEANERs to atleast one per chunk with a minimum of two. > have a nice day. > > ------=_NextPart_000_001B_01BF5CF2.35C4E380 > Content-Type: application/octet-stream; > name="onconfig" > Content-Transfer-Encoding: quoted-printable > Content-Disposition: attachment; > filename="onconfig" > > #************************************************************************= > **=0A= > #=0A= > # INFORMIX SOFTWARE, INC.=0A= > #=0A= > # Title: onconfig.std=0A= > # Description: Informix Dynamic Server Configuration Parameters=0A= > #=0A= > #************************************************************************= > **=0A= > =0A= > # Root Dbspace Configuration=0A= > =0A= > ROOTNAME rootdbs # Root dbspace name=0A= > ROOTPATH /info1/rootdbs.000 # Path for device containing root = > dbspace=0A= > ROOTOFFSET 0 # Offset of root dbspace into device = > (Kbytes)=0A= > ROOTSIZE 50000 # Size of root dbspace (Kbytes)=0A= > =0A=> # Disk Mirroring Configuration Parameters=0A= > =0A= > MIRROR 0 # Mirroring flag (Yes =3D 1, No =3D 0)=0A= > MIRRORPATH # Path for device containing mirrored = > root=0A= > MIRROROFFSET 0 # Offset into mirrored device (Kbytes)=0A= > =0A=> # Physical Log Configuration=0A= > =0A= > PHYSDBS rootdbs # Location (dbspace) of physical log=0A= > PHYSFILE 1000 # Physical log file size (Kbytes)=0A= > =0A=> # Logical Log Configuration=0A= > =0A= > LOGFILES 6 # Number of logical log files=0A= > LOGSIZE 500 # Logical log size (Kbytes)=0A= > =0A= > # Diagnostics=0A= > =0A=> MSGPATH /opt/informix/online.log # System message log file path=0A= > CONSOLE /dev/null # System console message path=0A= > ALARMPROGRAM /opt/informix/etc/log_full.sh # Alarm program path=0A= > SYSALARMPROGRAM /opt/informix/etc/evidence.sh # System Alarm program path=0A= > TBLSPACE_STATS 1 =0A= > =0A=> # System Archive Tape Device=0A= > =0A= > TAPEDEV /dev/null # Tape device path =0A= > TAPEBLK 16 # Tape block size (Kbytes)=0A= > TAPESIZE 10240 # Maximum amount of data to put on tape = > (Kbytes)=0A= > =0A= > # Log Archive Tape Device=0A= > =0A= > LTAPEDEV /dev/null # Log tape device path=0A= > LTAPEBLK 16 # Log tape block size (Kbytes)=0A= > LTAPESIZE 10240 # Max amount of data to put on log tape = > (Kbytes)=0A= > =0A= > # Optical=0A= > =0A=> STAGEBLOB # Informix Dynamic Server/Optical = > staging area=0A= > =0A= > # System Configuration=0A= > =0A= > SERVERNUM 0 # Unique id corresponding to a Dynamic = > Server instance=0A= > DBSERVERNAME ol_linuxshm # Name of default database server=0A= > DBSERVERALIASES ol_linux # List of alternate dbservernames=0A= > NETTYPE ipcshm,1,10,CPU # Configure poll thread(s) for nettype=0A= > NETTYPE soctcp,1,10,NET # Configure poll thread(s) for nettype=0A= > DEADLOCK_TIMEOUT 60 # Max time to wait of lock in = > distributed env.=0A= > RESIDENT 0 # Forced residency flag (Yes =3D 1, No = > =3D 0)=0A= > =0A= > MULTIPROCESSOR 0 # 0 for single-processor, 1 for = > multi-processor=0A= > NUMCPUVPS 1 # Number of user (cpu) vps=0A= > SINGLE_CPU_VP 0 # If non-zero, limit number of cpu vps = > to one=0A= > =0A=> #NOAGE 0 # Process aging=0A= > #AFF_SPROC 0 # Affinity start processor=0A= > #AFF_NPROCS 0 # Affinity number of processors=0A= > =0A= > # Shared Memory Parameters=0A= > =0A= > LOCKS 2000 # Maximum number of locks=0A= > BUFFERS 10000 # Maximum number of shared buffers=0A= > NUMAIOVPS # Number of IO vps=0A= > PHYSBUFF 32 # Physical log buffer size (Kbytes)=0A= > LOGBUFF 32 # Logical log b