Informix Warehouse ACcelerator : Problem with the

Hello, I'm testing the new product Informix Warehouse Accelerator. I created a data mart (estimated size :1 Gb) When I want to load this data mart, I have the following error : The mart DM1 can not be stored because the memory of accelerator CORSICA is not sufficient for a mart of that size. Additional diagnostic information: 'The processing of the DWA command was aborted due to an error on severity level 8 with diagnostic information 'The mart "DM1" is too big to fit into memory of the accelerator. Before loading the mart, "32" MB of shared memory were available on the failing node; total mart size is estimated at "301" MB per node. "Memory exhausted in stage 'compressing "informix"."agec"(3 of 3)', processed close to 2% of table data"'.' (AQT10213I) How to grow the shared memory for my node ? Indeed, my server's physical memory is 10 GB. I think that my dwainst.conf is correctly configured. Normally, I can load my data mart to IWA server with my configuration. My dwainst.conf is the following : # DWA storage directory # Stores the catalog, marts, logs, traces etc. DWADIR=/home/dwa/ # Starting port number # Grows to START_PORT-1+DEFAULT_NUMBER_ENTITLED_NODES*4 . START_PORT=21020 # Number of DWA_CM processes NUM_NODES=2 # Worker shared memory # SHM (in Megabyte) for all worker nodes. # Minimum value is 1 percent of physical memory. WORKER_SHM=7000 # Coordinator shared memory # SHM (in Megabyte) for all coordinator nodes. # Minimum value is 1 percent of physical memory. COORDINATOR_SHM=1000 # DRDA network interface # For running the accelerator on a separate computer ask the system # administrator for the network interface (e.g. eth0) that the Informix # database server should connect to or run the ifconfig command to find the # network interface. DRDA_INTERFACE=eth0 -- Franck Thomas ConsultiX franck.thomas@consult-ix.fr Téléphone : 33 (0) 1 39 12 18 00 Mobile : 33 (0) 6 78 81 09 33 Fax : 33 (0) 1 39 12 18 18

Hello. Mr Martin Fu has published several tutorials on his forum, check it out: https://www.ibm.com/developerworks/mydeveloperworks/blogs/2fa81a5c-cb30-4873-b77 5-1370151e3614/entry/informix_warehouse_accelerator_configuration_tips_5_size_of _dev_shm3?lang=en Hope it helps. Best regards. Em 05/05/2011 08:31, Franck THOMAS escreveu: > Hello, > > I'm testing the new product Informix Warehouse Accelerator. > I created a data mart (estimated size :1 Gb) > > When I want to load this data mart, I have the following error : > The mart DM1 can not be stored because the memory of accelerator CORSICA > is not sufficient for a mart of that size. Additional diagnostic > information: 'The processing of the DWA command was aborted due to an > error on severity level 8 with diagnostic information 'The mart "DM1" is > too big to fit into memory of the accelerator. Before loading the mart, > "32" MB of shared memory were available on the failing node; total mart > size is estimated at "301" MB per node. "Memory exhausted in stage > 'compressing "informix"."agec"(3 of 3)', processed close to 2% of table > data"'.' (AQT10213I) > > How to grow the shared memory for my node ? > > Indeed, my server's physical memory is 10 GB. > I think that my dwainst.conf is correctly configured. > > Normally, I can load my data mart to IWA server with my configuration. > > My dwainst.conf is the following : > > # DWA storage directory > # Stores the catalog, marts, logs, traces etc. > DWADIR=/home/dwa/ > > # Starting port number > # Grows to START_PORT-1+DEFAULT_NUMBER_ENTITLED_NODES*4 . > START_PORT=21020 > > # Number of DWA_CM processes > NUM_NODES=2 > > # Worker shared memory > # SHM (in Megabyte) for all worker nodes. > # Minimum value is 1 percent of physical memory. > WORKER_SHM=7000 > > # Coordinator shared memory > # SHM (in Megabyte) for all coordinator nodes. > # Minimum value is 1 percent of physical memory. > COORDINATOR_SHM=1000 > > # DRDA network interface > # For running the accelerator on a separate computer ask the system > # administrator for the network interface (e.g. eth0) that the Informix > # database server should connect to or run the ifconfig command to find the > # network interface. > DRDA_INTERFACE=eth0 > -- Alexandre Marini Tecnologia da Informação - DBA msn: alexandre_marini@hotmail.com SEFAZ-MS / SGI-UGSR / Sistemas IBM-Informix Cert-Info-Mgmt_color <Cert-Info-Mgmt_color.jpg> IBM Certified System Administrator - Informix Dynamic Server V10 / V11 See you at <http://www.iiug.org/conf/2011/iiug/>;

Frank, Please check this link https://www.ibm.com/developerworks/mydeveloperworks/blogs/2fa81a5c-cb30= -4873-b775-1370151e3614/entry/informix_warehouse_accelerator_configurat= ion_tips_3_linux_kernel4?lang=3Den_us Some of your Linux Kernel parameters are too low. Kind regards Sandor -----------------------------------------------------------------------= -------------------------------------------------------------------- IBM Deutschland Research & Development GmbH / Vorsitzender des Aufsichtsrats: Martin Jetter Gesch=E4ftsf=FChrung: Dirk Wittkopp Sitz der Gesellschaft: B=F6blingen / Registergericht: Amtsgericht Stutt= gart, HRB 243294 = "Franck THOMAS" = <franck.thomas@co = nsult-ix.fr> = To Sent by: ids@iiug.org = ids-bounces@iiug. = cc org = Subj= ect Informix Warehouse ACcelerator := 05/05/2011 02:31 Problem with .... [23603] = PM = = = Please respond to = ids@iiug.org = = = Hello, I'm testing the new product Informix Warehouse Accelerator. I created a data mart (estimated size :1 Gb) When I want to load this data mart, I have the following error : The mart DM1 can not be stored because the memory of accelerator CORSIC= A is not sufficient for a mart of that size. Additional diagnostic information: 'The processing of the DWA command was aborted due to an error on severity level 8 with diagnostic information 'The mart "DM1" i= s too big to fit into memory of the accelerator. Before loading the mart,= "32" MB of shared memory were available on the failing node; total mart= size is estimated at "301" MB per node. "Memory exhausted in stage 'compressing "informix"."agec"(3 of 3)', processed close to 2% of table= data"'.' (AQT10213I) How to grow the shared memory for my node ? Indeed, my server's physical memory is 10 GB. I think that my dwainst.conf is correctly configured. Normally, I can load my data mart to IWA server with my configuration. My dwainst.conf is the following : # DWA storage directory # Stores the catalog, marts, logs, traces etc. DWADIR=3D/home/dwa/ # Starting port number # Grows to START_PORT-1+DEFAULT_NUMBER_ENTITLED_NODES*4 . START_PORT=3D21020 # Number of DWA_CM processes NUM_NODES=3D2 # Worker shared memory # SHM (in Megabyte) for all worker nodes. # Minimum value is 1 percent of physical memory. WORKER_SHM=3D7000 # Coordinator shared memory # SHM (in Megabyte) for all coordinator nodes. # Minimum value is 1 percent of physical memory. COORDINATOR_SHM=3D1000 # DRDA network interface # For running the accelerator on a separate computer ask the system # administrator for the network interface (e.g. eth0) that the Informix= # database server should connect to or run the ifconfig command to find= the # network interface. DRDA_INTERFACE=3Deth0 -- Franck Thomas ConsultiX franck.thomas@consult-ix.fr T=E9l=E9phone : 33 (0) 1 39 12 18 00 Mobile : 33 (0) 6 78 81 09 33 Fax : 33 (0) 1 39 12 18 18 ***********************************************************************= ******** Forum Note: Use "Reply" to post a response in the discussion forum. =

Hi, probably the worker node has experienced the shortage of memory. Which means that most probably there are already other data marts existing in the accelerator. You can check the size of available shared memory space (at best before you create the new data mart) by looking at the accelerator instance's "Properties" view in the ISAO Studio GUI. You should find something saying "Shared Memory available: x GB of y GB". Here y is the total (i.e. ~7 GB). The difference between the two numbers is what is already used (by other data marts). During the loading of a data mart additional memory is needed, e.g. for the compression processing. This is memory allocated "privately" by a node and thus not part of the shared memory. But still it can cause a memory shortage (of overall physically available memory versus physical memory already allocated and in use). Other than that, as already pointed out, also check the size and space available of /dev/shm. It doesn't seem to me, that the bottleneck is in the coordinator node. Please let me know your findings according to the above. Regards, Martin -- Martin Fuerderer IBM Informix Development Munich, Germany Information Management Read about the Informix Warehouse Accelerator: https://www.ibm.com/developerworks/mydeveloperworks/ blogs/2fa81a5c-cb30-4873-b775-1370151e3614/entry/ introducing_informix_warehouse_accelerator9?lang=en IBM Deutschland Research & Development GmbH Chairman of the Supervisory Board: Martin Jetter Board of Management: Dirk Wittkopp Corporate Seat: Boeblingen, Germany Reg.-Gericht: Amtsgericht Stuttgart, HRB 243294 ids-bounces@iiug.org wrote on 05/05/2011 02:31:46 PM: > > Hello, > > I'm testing the new product Informix Warehouse Accelerator. > I created a data mart (estimated size :1 Gb) > > When I want to load this data mart, I have the following error : > The mart DM1 can not be stored because the memory of accelerator CORSICA > is not sufficient for a mart of that size. Additional diagnostic > information: 'The processing of the DWA command was aborted due to an > error on severity level 8 with diagnostic information 'The mart "DM1" is > too big to fit into memory of the accelerator. Before loading the mart, > "32" MB of shared memory were available on the failing node; total mart > size is estimated at "301" MB per node. "Memory exhausted in stage > 'compressing "informix"."agec"(3 of 3)', processed close to 2% of table > data"'.' (AQT10213I) > > How to grow the shared memory for my node ? > > Indeed, my server's physical memory is 10 GB. > I think that my dwainst.conf is correctly configured. > > Normally, I can load my data mart to IWA server with my configuration. > > My dwainst.conf is the following : > > # DWA storage directory > # Stores the catalog, marts, logs, traces etc. > DWADIR=/home/dwa/ > > # Starting port number > # Grows to START_PORT-1+DEFAULT_NUMBER_ENTITLED_NODES*4 . > START_PORT=21020 > > # Number of DWA_CM processes > NUM_NODES=2 > > # Worker shared memory > # SHM (in Megabyte) for all worker nodes. > # Minimum value is 1 percent of physical memory. > WORKER_SHM=7000 > > # Coordinator shared memory > # SHM (in Megabyte) for all coordinator nodes. > # Minimum value is 1 percent of physical memory. > COORDINATOR_SHM=1000 > > # DRDA network interface > # For running the accelerator on a separate computer ask the system > # administrator for the network interface (e.g. eth0) that the Informix > # database server should connect to or run the ifconfig command to find the > # network interface. > DRDA_INTERFACE=eth0 > > -- > Franck Thomas > ConsultiX > franck.thomas@consult-ix.fr > Téléphone : 33 (0) 1 39 12 18 00 > Mobile : 33 (0) 6 78 81 09 33 > Fax : 33 (0) 1 39 12 18 18 > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. >

Hi, I have only one data mart. I found why I cannot load my data mart, my SHMMAX is too little on my server. I changed my SHMMAX parameter to 1 024 Mb instead of 330 Mb and now I can load my data mart. Thank you for your links and your explanations because they are very helpful. Le 05/05/2011 15:58, Martin Fuerderer a écrit : > Hi, > > probably the worker node has experienced the > shortage of memory. Which means that most > probably there are already other data marts > existing in the accelerator. You can check the > size of available shared memory space (at best > before you create the new data mart) by looking > at the accelerator instance's "Properties" view in > the ISAO Studio GUI. You should find something > saying "Shared Memory available: x GB of y GB". > Here y is the total (i.e. ~7 GB). The difference > between the two numbers is what is already used > (by other data marts). > > During the loading of a data mart additional memory > is needed, e.g. for the compression processing. This > is memory allocated "privately" by a node and thus > not part of the shared memory. But still it can cause > a memory shortage (of overall physically available > memory versus physical memory already allocated > and in use). > > Other than that, as already pointed out, also check > the size and space available of /dev/shm. > > It doesn't seem to me, that the bottleneck is in the > coordinator node. > > Please let me know your findings according to the above. > > Regards, Martin -- Franck Thomas ConsultiX franck.thomas@consult-ix.fr Téléphone : 33 (0) 1 39 12 18 00 Mobile : 33 (0) 6 78 81 09 33 Fax : 33 (0) 1 39 12 18 18