Moving LARGE amount of data online

Hi, RedHat EL4, 11.50.UC5 Big question here - I have ~1TB of data spread over ~450 2GB chunks in the engine. There are another 1000 chunks also in the engine - total is 1500 chunks. I need to move these 450 chunks to a different disk subsystem (they need to move from fast disk to slow disk to make room for more data). Really easy if I shut down the engine (shut down, dd the files, move the sym links, start up engine) - but I can't have the engine offline that long. So how do you move 1TB of data across 450 chunks without taking down the engine? I can do it through mirroring (Create a mirror, take the primary offline, move the primary symlink, bring the primary back online and drop the mirror). This moves the data twice though (1. to the mirror, 2. back to the primary). Does anyone know how to make the mirror chunk the primary? Then I can just create the mirrors, switch the primary and mirror (both online) and then drop the new mirrors (original primaries). Any other ideas to move this data without shutting down the engine? The chunk data can be unavailable during this time - just the other 1000 chunks have to be up. Thanks Jarrod Teale Team Lead - Manufacturing Execution Systems Automation & Process Control Group NZ Technical Fonterra Fonterra Extn: 77525 DDI: +64 7 850 7525 Mobile: +64 21 968 364 fax: +64 7 849 7855 email: jarrod.teale@fonterra.com <mailto:jarrod.teale@fonterra.com> DISCLAIMER: This email contains confidential information and may be legally privileged. If you are not the intended recipient or have received this email in error, please notify the sender immediately and destroy this email. You may not use, disclose or copy this email or its attachments in any way. Any opinions expressed in this email are those of the author and are not necessarily those of the Fonterra Co-operative Group. http://www.fonterra.com/

Is the ~450 2GB chunks stored in a SAN, if so then talk your SAN administrator. They might be able to set up mirroring behind the scene to mirror from faster disk subsystem to slower disk subsystem. Once the mirror (LUNs) are in sync, they will be able to break the mirroring. They will be able to make the mirror the primary. From the Informix stand point, you don't even have to change the symbolic link. Caution: Your existing SAN solution might not have this feature. There are lot of tools available in the market from different vendors IBM/EMC/HP. On a side note, have you thought about consolidating data in 2GB chunks to larger chunks. Thank you, Kannan -----Original Message----- From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Jarrod Teale Sent: Friday, October 30, 2009 5:47 AM To: ids@iiug.org Subject: Moving LARGE amount of data online [17839] Hi, RedHat EL4, 11.50.UC5 Big question here - I have ~1TB of data spread over ~450 2GB chunks in the engine. There are another 1000 chunks also in the engine - total is 1500 chunks. I need to move these 450 chunks to a different disk subsystem (they need to move from fast disk to slow disk to make room for more data). Really easy if I shut down the engine (shut down, dd the files, move the sym links, start up engine) - but I can't have the engine offline that long. So how do you move 1TB of data across 450 chunks without taking down the engine? I can do it through mirroring (Create a mirror, take the primary offline, move the primary symlink, bring the primary back online and drop the mirror). This moves the data twice though (1. to the mirror, 2. back to the primary). Does anyone know how to make the mirror chunk the primary? Then I can just create the mirrors, switch the primary and mirror (both online) and then drop the new mirrors (original primaries). Any other ideas to move this data without shutting down the engine? The chunk data can be unavailable during this time - just the other 1000 chunks have to be up. Thanks Jarrod Teale Team Lead - Manufacturing Execution Systems Automation & Process Control Group NZ Technical Fonterra Fonterra Extn: 77525 DDI: +64 7 850 7525 Mobile: +64 21 968 364 fax: +64 7 849 7855 email: jarrod.teale@fonterra.com <mailto:jarrod.teale@fonterra.com> DISCLAIMER: This email contains confidential information and may be legally privileged. If you are not the intended recipient or have received this email in error, please notify the sender immediately and destroy this email. You may not use, disclose or copy this email or its attachments in any way. Any opinions expressed in this email are those of the author and are not necessarily those of the Fonterra Co-operative Group. http://www.fonterra.com/ ************************************************************************ ******* Forum Note: Use "Reply" to post a response in the discussion forum.

Hi Jarrod, If you don't have SAN setup as nice as Kannan suggested with the ability to split your mirrors/etc at disk subsystem level.... What about fragmentation?.... i.e. set up your new chunks on slow disk, alter your current chunks (tables in dbspaces on those chunks) to round robin and/or frag strategy into the new chunks on slow disk. I forgot if that process means the data is not accessible during the fragmentation process (but you said unavailable was OK). I also don't remember how fast the process is (you didn't specify a speed requirement)... You might also look into doing something creative with onbar backup/restore... Sounds like fun... I'm jealous... NJ -----Original Message----- From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Jarrod Teale Sent: Friday, October 30, 2009 4:47 AM To: ids@iiug.org Subject: Moving LARGE amount of data online [17839] Hi, RedHat EL4, 11.50.UC5 Big question here - I have ~1TB of data spread over ~450 2GB chunks in the engine. There are another 1000 chunks also in the engine - total is 1500 chunks. I need to move these 450 chunks to a different disk subsystem (they need to move from fast disk to slow disk to make room for more data). Really easy if I shut down the engine (shut down, dd the files, move the sym links, start up engine) - but I can't have the engine offline that long. So how do you move 1TB of data across 450 chunks without taking down the engine? I can do it through mirroring (Create a mirror, take the primary offline, move the primary symlink, bring the primary back online and drop the mirror). This moves the data twice though (1. to the mirror, 2. back to the primary). Does anyone know how to make the mirror chunk the primary? Then I can just create the mirrors, switch the primary and mirror (both online) and then drop the new mirrors (original primaries). Any other ideas to move this data without shutting down the engine? The chunk data can be unavailable during this time - just the other 1000 chunks have to be up. Thanks Jarrod Teale Team Lead - Manufacturing Execution Systems Automation & Process Control Group NZ Technical Fonterra Fonterra Extn: 77525 DDI: +64 7 850 7525 Mobile: +64 21 968 364 fax: +64 7 849 7855 email: jarrod.teale@fonterra.com <mailto:jarrod.teale@fonterra.com> DISCLAIMER: This email contains confidential information and may be legally privileged. If you are not the intended recipient or have received this email in error, please notify the sender immediately and destroy this email. You may not use, disclose or copy this email or its attachments in any way. Any opinions expressed in this email are those of the author and are not necessarily those of the Fonterra Co-operative Group. http://www.fonterra.com/ ************************************************************************ ******* Forum Note: Use "Reply" to post a response in the discussion forum.

Hi I enhance the option suggested by Nancy . Some years ago, I need to move all chunks of an instance from a Storage disk Shark , to new disks in a DS8000 storage. In that time I had only 1 fiber channel and I could not have both storages connected to the server at same time. My decision (Online applications) was: The instance had ~ 100 GB , I had disk space in internal disks of the server more than 100GB I created "mirrors" raw's for each chunk. Change OnConfig to MIRROR Yes - Started mirror all dbspaces (each chunk of dbspaces) - After all dbspaces mirrored I put down the Chunk Principal and the system continues running on mirrors - Drop all raw's from Shark disks - Create new raw's on DS8000 (after change fiber connection from shark to ds8000) - Starts to put chunks Principal UP again in DS8000 All the task runned well ... Versions: IDS 7.31.UD5 OS: AIX 5.1 Best regards R Ferronato > To: ids@iiug.org > From: nsebastian@flinnsci.com > Subject: RE: Moving LARGE amount of data online [17843] > Date: Fri, 30 Oct 2009 08:01:17 -0400 > > Hi Jarrod, > If you don't have SAN setup as nice as Kannan suggested with the ability > to split your mirrors/etc at disk subsystem level.... > > What about fragmentation?.... > i.e. set up your new chunks on slow disk, alter your current chunks > (tables in dbspaces on those chunks) to round robin and/or frag strategy > into the new chunks on slow disk. > I forgot if that process means the data is not accessible during the > fragmentation process (but you said unavailable was OK). I also don't > remember how fast the process is (you didn't specify a speed > requirement)... > > You might also look into doing something creative with onbar > backup/restore... > Sounds like fun... I'm jealous... > NJ > > -----Original Message----- > From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of > Jarrod Teale > Sent: Friday, October 30, 2009 4:47 AM > To: ids@iiug.org > Subject: Moving LARGE amount of data online [17839] > > Hi, > RedHat EL4, 11.50.UC5 > Big question here - I have ~1TB of data spread over ~450 2GB chunks in > the engine. > There are another 1000 chunks also in the engine - total is 1500 chunks. > > I need to move these 450 chunks to a different disk subsystem (they need > > to move from fast disk to slow disk to make room for more data). > Really easy if I shut down the engine (shut down, dd the files, move the > > sym links, start up engine) - but I can't have the engine offline that > long. > > So how do you move 1TB of data across 450 chunks without taking down the > > engine? > I can do it through mirroring (Create a mirror, take the primary > offline, move the primary symlink, bring the primary back online and > drop the mirror). > This moves the data twice though (1. to the mirror, 2. back to the > primary). > > Does anyone know how to make the mirror chunk the primary? Then I can > just create the mirrors, switch the primary and mirror (both online) and > > then drop the new mirrors (original primaries). > > Any other ideas to move this data without shutting down the engine? The > chunk data can be unavailable during this time - just the other 1000 > chunks have to be up. > > Thanks > > Jarrod Teale > > Team Lead - Manufacturing Execution Systems > > Automation & Process Control Group > > NZ Technical > > Fonterra > > Fonterra Extn: 77525 > > DDI: +64 7 850 7525 > > Mobile: +64 21 968 364 > fax: +64 7 849 7855 > email: jarrod.teale@fonterra.com > <mailto:jarrod.teale@fonterra.com> > > DISCLAIMER: > This email contains confidential information and may be legally > privileged. If > you are not the intended recipient or have received this email in error, > > please notify the sender immediately and destroy this email. > You may not use, disclose or copy this email or its attachments in any > way. > Any opinions expressed in this email are those of the author and are not > > necessarily those of the Fonterra Co-operative Group. > http://www.fonterra.com/ > > ************************************************************************ > ******* > Forum Note: Use "Reply" to post a response in the discussion forum. > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > _________________________________________________________________ Windows Live: Friends get your Flickr, Yelp, and Digg updates when they e-mail you. http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/social- network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:en-xm:SI_SB_3:092010

If you can take a brief outage, you can mirror the chunks onto slower storage, bring down the server, reverse the symlinks between the primary and mirror chunks, restart the server and drop the mirrors. Note that you have IDS 11.50xC5 for which you can license compression. Getting that enabled you can compress the data in-place without any downtime and at least double the amount of data you can keep in the existing chunks! You should experience a performance gain to boot! Art Art S. Kagel Oninit (www.oninit.com) IIUG Board of Directors (art@iiug.org) Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Oninit, the IIUG, nor any other organization with which I am associated either explicitly or implicitly. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Fri, Oct 30, 2009 at 5:46 AM, Jarrod Teale <Jarrod.Teale@fonterra.com>wrote: > Hi, > RedHat EL4, 11.50.UC5 > Big question here - I have ~1TB of data spread over ~450 2GB chunks in > the engine. > There are another 1000 chunks also in the engine - total is 1500 chunks. > > I need to move these 450 chunks to a different disk subsystem (they need > to move from fast disk to slow disk to make room for more data). > Really easy if I shut down the engine (shut down, dd the files, move the > sym links, start up engine) - but I can't have the engine offline that > long. > > So how do you move 1TB of data across 450 chunks without taking down the > engine? > I can do it through mirroring (Create a mirror, take the primary > offline, move the primary symlink, bring the primary back online and > drop the mirror). > This moves the data twice though (1. to the mirror, 2. back to the > primary). > > Does anyone know how to make the mirror chunk the primary? Then I can > just create the mirrors, switch the primary and mirror (both online) and > then drop the new mirrors (original primaries). > > Any other ideas to move this data without shutting down the engine? The > chunk data can be unavailable during this time - just the other 1000 > chunks have to be up. > > Thanks > > Jarrod Teale > > Team Lead - Manufacturing Execution Systems > > Automation & Process Control Group > > NZ Technical > > Fonterra > > Fonterra Extn: 77525 > > DDI: +64 7 850 7525 > > Mobile: +64 21 968 364 > fax: +64 7 849 7855 > email: jarrod.teale@fonterra.com > <mailto:jarrod.teale@fonterra.com> > > DISCLAIMER: > This email contains confidential information and may be legally privileged. > If > you are not the intended recipient or have received this email in error, > please notify the sender immediately and destroy this email. > You may not use, disclose or copy this email or its attachments in any way. > Any opinions expressed in this email are those of the author and are not > necessarily those of the Fonterra Co-operative Group. > http://www.fonterra.com/ > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --00151747befc6c084304772806f4