Re: [U2] Huge Dynamic Unidata file
You are of course right - in my defence it was a few years ago ;) -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: 26 April 2012 12:21 To: 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file I thought TMPPATH was not valid for Dynamic files: ?? The TMPPATH option is invalid if any DYNAMIC options are specified (or if the starting file is dynamic and no file type options are specified). -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Symeon Breen Sent: Thursday, April 26, 2012 6:18 AM To: 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file That's is pretty big, my personal experience with big files on udt was up to about 60Gig - we did use memresize no problems - but had to set the TMPPATH to another drive. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: 26 April 2012 12:00 To: 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file Indexes and all, 91 G with about 75 million records. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Symeon Breen Sent: Thursday, April 26, 2012 4:08 AM To: 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file Out of interest, how big is this file and how many records ? -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: 25 April 2012 21:30 To: 'U2 Users List'; jonathan.lec...@blairswindows.co.uk Subject: Re: [U2] Huge Dynamic Unidata file I think at the time I wrote this, Unidata 6 I think, I could not even get memresize to handle overflow that was larger than 2 gigabytes. So I thought the solution by Unidata was to create an overflow file for every dat segment. I have never specified the OVERFLOW option when using memresize. Plus, I rarely have enough space in a file system to handle 2 copies of one of my big file. One work around for the file system space issue is to symbolically link the original file in another file system. But that is a lot of work when you have a lot of dat files. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of dean.armbrus...@ferguson.com Sent: Wednesday, April 25, 2012 3:16 PM To: u2-users@listserver.u2ug.org; jonathan.lec...@blairswindows.co.uk Subject: Re: [U2] Huge Dynamic Unidata file Were you using the OVERFLOW option with memresize? If not, memresize should not be creating the extra over files. If memresize did create the extra over files without the OVERFLOW option, then that would be a bug in memresize. Dean Armbruster System Analyst 757-989-2839 -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: Wednesday, April 25, 2012 12:29 To: 'Jonathan Leckie'; 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file I resize most of my Dynamic files this way. I don't like having a small overxxx segment for every datxxx segment that memresize creates.. By creating the new file my self, I don't have a lot of these small overxxx segments that are never used. I also wrote a process to Select the old file and create a SAVELIST. Without going into to much detail, the process uses PHANTOM to spawn off a number of Unidata copies. So there are a number of simultaneous processes working to build the new file. Each PHANTOM Copy knows how many total phantoms are working on the file and what sequence IT is within the total number of PHANTOMS. Each Phantom handles part of the SAVELIST. Each PHANTOM does the iteration below until the list is exhausted: 1. QSELECT SAVEDLISTS listname000. (process 2 would use 001, process 3 would use 002 etc) 2. COPY FROM old.file TO new.file 3. Increment the list counter from 000 by the number of Phantoms and go back to step one to process a new segment of the savelist. 4. If a process cannot find the next savelist segment in SAVEDLISTS, it is done. This process is almost as fast as memresize. You can control what file systems are used for space reasons and you don't get scads of overxxx segments in your FILE.NAME directory (one of my dynamic files has 39 dat segments and just a over001). So I don't use memresize for dynamic files anymore. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Jonathan Leckie Sent: Wednesday, April 25, 2012 9:36 AM To: u2-users@listserve
Re: [U2] Huge Dynamic Unidata file
I thought TMPPATH was not valid for Dynamic files: The TMPPATH option is invalid if any DYNAMIC options are specified (or if the starting file is dynamic and no file type options are specified). -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Symeon Breen Sent: Thursday, April 26, 2012 6:18 AM To: 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file That's is pretty big, my personal experience with big files on udt was up to about 60Gig - we did use memresize no problems - but had to set the TMPPATH to another drive. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: 26 April 2012 12:00 To: 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file Indexes and all, 91 G with about 75 million records. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Symeon Breen Sent: Thursday, April 26, 2012 4:08 AM To: 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file Out of interest, how big is this file and how many records ? -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: 25 April 2012 21:30 To: 'U2 Users List'; jonathan.lec...@blairswindows.co.uk Subject: Re: [U2] Huge Dynamic Unidata file I think at the time I wrote this, Unidata 6 I think, I could not even get memresize to handle overflow that was larger than 2 gigabytes. So I thought the solution by Unidata was to create an overflow file for every dat segment. I have never specified the OVERFLOW option when using memresize. Plus, I rarely have enough space in a file system to handle 2 copies of one of my big file. One work around for the file system space issue is to symbolically link the original file in another file system. But that is a lot of work when you have a lot of dat files. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of dean.armbrus...@ferguson.com Sent: Wednesday, April 25, 2012 3:16 PM To: u2-users@listserver.u2ug.org; jonathan.lec...@blairswindows.co.uk Subject: Re: [U2] Huge Dynamic Unidata file Were you using the OVERFLOW option with memresize? If not, memresize should not be creating the extra over files. If memresize did create the extra over files without the OVERFLOW option, then that would be a bug in memresize. Dean Armbruster System Analyst 757-989-2839 -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: Wednesday, April 25, 2012 12:29 To: 'Jonathan Leckie'; 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file I resize most of my Dynamic files this way. I don't like having a small overxxx segment for every datxxx segment that memresize creates.. By creating the new file my self, I don't have a lot of these small overxxx segments that are never used. I also wrote a process to Select the old file and create a SAVELIST. Without going into to much detail, the process uses PHANTOM to spawn off a number of Unidata copies. So there are a number of simultaneous processes working to build the new file. Each PHANTOM Copy knows how many total phantoms are working on the file and what sequence IT is within the total number of PHANTOMS. Each Phantom handles part of the SAVELIST. Each PHANTOM does the iteration below until the list is exhausted: 1. QSELECT SAVEDLISTS listname000. (process 2 would use 001, process 3 would use 002 etc) 2. COPY FROM old.file TO new.file 3. Increment the list counter from 000 by the number of Phantoms and go back to step one to process a new segment of the savelist. 4. If a process cannot find the next savelist segment in SAVEDLISTS, it is done. This process is almost as fast as memresize. You can control what file systems are used for space reasons and you don't get scads of overxxx segments in your FILE.NAME directory (one of my dynamic files has 39 dat segments and just a over001). So I don't use memresize for dynamic files anymore. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Jonathan Leckie Sent: Wednesday, April 25, 2012 9:36 AM To: u2-users@listserver.u2ug.org Subject: [U2] Huge Dynamic Unidata file I have a very large file that I don't have enough free space to memresize, however howabout I create new dynamic (temporary) file and then copy all the records (in ECL) to the new file and then (unix) copy the temporary file over the
Re: [U2] Huge Dynamic Unidata file
That's is pretty big, my personal experience with big files on udt was up to about 60Gig - we did use memresize no problems - but had to set the TMPPATH to another drive. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: 26 April 2012 12:00 To: 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file Indexes and all, 91 G with about 75 million records. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Symeon Breen Sent: Thursday, April 26, 2012 4:08 AM To: 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file Out of interest, how big is this file and how many records ? -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: 25 April 2012 21:30 To: 'U2 Users List'; jonathan.lec...@blairswindows.co.uk Subject: Re: [U2] Huge Dynamic Unidata file I think at the time I wrote this, Unidata 6 I think, I could not even get memresize to handle overflow that was larger than 2 gigabytes. So I thought the solution by Unidata was to create an overflow file for every dat segment. I have never specified the OVERFLOW option when using memresize. Plus, I rarely have enough space in a file system to handle 2 copies of one of my big file. One work around for the file system space issue is to symbolically link the original file in another file system. But that is a lot of work when you have a lot of dat files. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of dean.armbrus...@ferguson.com Sent: Wednesday, April 25, 2012 3:16 PM To: u2-users@listserver.u2ug.org; jonathan.lec...@blairswindows.co.uk Subject: Re: [U2] Huge Dynamic Unidata file Were you using the OVERFLOW option with memresize? If not, memresize should not be creating the extra over files. If memresize did create the extra over files without the OVERFLOW option, then that would be a bug in memresize. Dean Armbruster System Analyst 757-989-2839 -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: Wednesday, April 25, 2012 12:29 To: 'Jonathan Leckie'; 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file I resize most of my Dynamic files this way. I don't like having a small overxxx segment for every datxxx segment that memresize creates.. By creating the new file my self, I don't have a lot of these small overxxx segments that are never used. I also wrote a process to Select the old file and create a SAVELIST. Without going into to much detail, the process uses PHANTOM to spawn off a number of Unidata copies. So there are a number of simultaneous processes working to build the new file. Each PHANTOM Copy knows how many total phantoms are working on the file and what sequence IT is within the total number of PHANTOMS. Each Phantom handles part of the SAVELIST. Each PHANTOM does the iteration below until the list is exhausted: 1. QSELECT SAVEDLISTS listname000. (process 2 would use 001, process 3 would use 002 etc) 2. COPY FROM old.file TO new.file 3. Increment the list counter from 000 by the number of Phantoms and go back to step one to process a new segment of the savelist. 4. If a process cannot find the next savelist segment in SAVEDLISTS, it is done. This process is almost as fast as memresize. You can control what file systems are used for space reasons and you don't get scads of overxxx segments in your FILE.NAME directory (one of my dynamic files has 39 dat segments and just a over001). So I don't use memresize for dynamic files anymore. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Jonathan Leckie Sent: Wednesday, April 25, 2012 9:36 AM To: u2-users@listserver.u2ug.org Subject: [U2] Huge Dynamic Unidata file I have a very large file that I don't have enough free space to memresize, however howabout I create new dynamic (temporary) file and then copy all the records (in ECL) to the new file and then (unix) copy the temporary file over the top of the original. Does that seem like a sensible idea? Regards Jonathan Leckie ** * This message has been scanned for viruses and dangerous content * and is believed to be clean. * * This email and any files transmitted with it are confidential and * intended solely for the use of the individual or entity to whom they * are addressed. * * If you have received this email in error please notify us at Blairs, * details can be found on our
Re: [U2] Huge Dynamic Unidata file
Indexes and all, 91 G with about 75 million records. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Symeon Breen Sent: Thursday, April 26, 2012 4:08 AM To: 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file Out of interest, how big is this file and how many records ? -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: 25 April 2012 21:30 To: 'U2 Users List'; jonathan.lec...@blairswindows.co.uk Subject: Re: [U2] Huge Dynamic Unidata file I think at the time I wrote this, Unidata 6 I think, I could not even get memresize to handle overflow that was larger than 2 gigabytes. So I thought the solution by Unidata was to create an overflow file for every dat segment. I have never specified the OVERFLOW option when using memresize. Plus, I rarely have enough space in a file system to handle 2 copies of one of my big file. One work around for the file system space issue is to symbolically link the original file in another file system. But that is a lot of work when you have a lot of dat files. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of dean.armbrus...@ferguson.com Sent: Wednesday, April 25, 2012 3:16 PM To: u2-users@listserver.u2ug.org; jonathan.lec...@blairswindows.co.uk Subject: Re: [U2] Huge Dynamic Unidata file Were you using the OVERFLOW option with memresize? If not, memresize should not be creating the extra over files. If memresize did create the extra over files without the OVERFLOW option, then that would be a bug in memresize. Dean Armbruster System Analyst 757-989-2839 -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: Wednesday, April 25, 2012 12:29 To: 'Jonathan Leckie'; 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file I resize most of my Dynamic files this way. I don't like having a small overxxx segment for every datxxx segment that memresize creates.. By creating the new file my self, I don't have a lot of these small overxxx segments that are never used. I also wrote a process to Select the old file and create a SAVELIST. Without going into to much detail, the process uses PHANTOM to spawn off a number of Unidata copies. So there are a number of simultaneous processes working to build the new file. Each PHANTOM Copy knows how many total phantoms are working on the file and what sequence IT is within the total number of PHANTOMS. Each Phantom handles part of the SAVELIST. Each PHANTOM does the iteration below until the list is exhausted: 1. QSELECT SAVEDLISTS listname000. (process 2 would use 001, process 3 would use 002 etc) 2. COPY FROM old.file TO new.file 3. Increment the list counter from 000 by the number of Phantoms and go back to step one to process a new segment of the savelist. 4. If a process cannot find the next savelist segment in SAVEDLISTS, it is done. This process is almost as fast as memresize. You can control what file systems are used for space reasons and you don't get scads of overxxx segments in your FILE.NAME directory (one of my dynamic files has 39 dat segments and just a over001). So I don't use memresize for dynamic files anymore. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Jonathan Leckie Sent: Wednesday, April 25, 2012 9:36 AM To: u2-users@listserver.u2ug.org Subject: [U2] Huge Dynamic Unidata file I have a very large file that I don't have enough free space to memresize, however howabout I create new dynamic (temporary) file and then copy all the records (in ECL) to the new file and then (unix) copy the temporary file over the top of the original. Does that seem like a sensible idea? Regards Jonathan Leckie ** * This message has been scanned for viruses and dangerous content * and is believed to be clean. * * This email and any files transmitted with it are confidential and * intended solely for the use of the individual or entity to whom they * are addressed. * * If you have received this email in error please notify us at Blairs, * details can be found on our website http://www.blairswindows.co.uk * * Name & Registered Office: * * Blairs Windows Limited * Registered office : 9 Baker Street, Greenock, PA15 4TU * Company No: SC393935, V.A.T. registration No: 108729111 ** ___ U2-Users mailing list U2-Users@lis
Re: [U2] Huge Dynamic Unidata file
Out of interest, how big is this file and how many records ? -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: 25 April 2012 21:30 To: 'U2 Users List'; jonathan.lec...@blairswindows.co.uk Subject: Re: [U2] Huge Dynamic Unidata file I think at the time I wrote this, Unidata 6 I think, I could not even get memresize to handle overflow that was larger than 2 gigabytes. So I thought the solution by Unidata was to create an overflow file for every dat segment. I have never specified the OVERFLOW option when using memresize. Plus, I rarely have enough space in a file system to handle 2 copies of one of my big file. One work around for the file system space issue is to symbolically link the original file in another file system. But that is a lot of work when you have a lot of dat files. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of dean.armbrus...@ferguson.com Sent: Wednesday, April 25, 2012 3:16 PM To: u2-users@listserver.u2ug.org; jonathan.lec...@blairswindows.co.uk Subject: Re: [U2] Huge Dynamic Unidata file Were you using the OVERFLOW option with memresize? If not, memresize should not be creating the extra over files. If memresize did create the extra over files without the OVERFLOW option, then that would be a bug in memresize. Dean Armbruster System Analyst 757-989-2839 -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: Wednesday, April 25, 2012 12:29 To: 'Jonathan Leckie'; 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file I resize most of my Dynamic files this way. I don't like having a small overxxx segment for every datxxx segment that memresize creates.. By creating the new file my self, I don't have a lot of these small overxxx segments that are never used. I also wrote a process to Select the old file and create a SAVELIST. Without going into to much detail, the process uses PHANTOM to spawn off a number of Unidata copies. So there are a number of simultaneous processes working to build the new file. Each PHANTOM Copy knows how many total phantoms are working on the file and what sequence IT is within the total number of PHANTOMS. Each Phantom handles part of the SAVELIST. Each PHANTOM does the iteration below until the list is exhausted: 1. QSELECT SAVEDLISTS listname000. (process 2 would use 001, process 3 would use 002 etc) 2. COPY FROM old.file TO new.file 3. Increment the list counter from 000 by the number of Phantoms and go back to step one to process a new segment of the savelist. 4. If a process cannot find the next savelist segment in SAVEDLISTS, it is done. This process is almost as fast as memresize. You can control what file systems are used for space reasons and you don't get scads of overxxx segments in your FILE.NAME directory (one of my dynamic files has 39 dat segments and just a over001). So I don't use memresize for dynamic files anymore. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Jonathan Leckie Sent: Wednesday, April 25, 2012 9:36 AM To: u2-users@listserver.u2ug.org Subject: [U2] Huge Dynamic Unidata file I have a very large file that I don't have enough free space to memresize, however howabout I create new dynamic (temporary) file and then copy all the records (in ECL) to the new file and then (unix) copy the temporary file over the top of the original. Does that seem like a sensible idea? Regards Jonathan Leckie ** * This message has been scanned for viruses and dangerous content * and is believed to be clean. * * This email and any files transmitted with it are confidential and * intended solely for the use of the individual or entity to whom they * are addressed. * * If you have received this email in error please notify us at Blairs, * details can be found on our website http://www.blairswindows.co.uk * * Name & Registered Office: * * Blairs Windows Limited * Registered office : 9 Baker Street, Greenock, PA15 4TU * Company No: SC393935, V.A.T. registration No: 108729111 ** ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users -- CONFIDENTIALITY NOTICE: If you have received this email in error, please immediately notify the sender by e-mail at the address shown. This email transmission may contain confidential information. This information is intended only for the use
Re: [U2] Huge Dynamic Unidata file
We are looking at options for larger files in UniData as there are some BIG data sets out there. The options that are being looked at include (and in no particular order): 1. 64 bit files - over 2Gb 2. Distributed files (like UniVerse) 3. More dat/over/idx files - going from datnnn to dat Each has pros and cons, and if there are other options let's hear them! If you have a need and a preference then please speak up. Regards JayJay On 25 Apr 2012, at 21:29, "Baakkonen, Rodney A (Rod) 46K" wrote: > >I think at the time I wrote this, Unidata 6 I think, I could not even get > memresize to handle overflow that was larger than 2 gigabytes. So I thought > the solution by Unidata was to create an overflow file for every dat segment. > I have never specified the OVERFLOW option when using memresize. Plus, I > rarely have enough space in a file system to handle 2 copies of one of my big > file. One work around for the file system space issue is to symbolically link > the original file in another file system. But that is a lot of work when you > have a lot of dat files. > > > > -Original Message- > From: u2-users-boun...@listserver.u2ug.org > [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of > dean.armbrus...@ferguson.com > Sent: Wednesday, April 25, 2012 3:16 PM > To: u2-users@listserver.u2ug.org; jonathan.lec...@blairswindows.co.uk > Subject: Re: [U2] Huge Dynamic Unidata file > > Were you using the OVERFLOW option with memresize? If not, memresize > should not be creating the extra over files. If memresize did create > the extra over files without the OVERFLOW option, then that would be a > bug in memresize. > > Dean Armbruster > System Analyst > 757-989-2839 > > -Original Message- > From: u2-users-boun...@listserver.u2ug.org > [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, > Rodney A (Rod) 46K > Sent: Wednesday, April 25, 2012 12:29 > To: 'Jonathan Leckie'; 'U2 Users List' > Subject: Re: [U2] Huge Dynamic Unidata file > > > I resize most of my Dynamic files this way. I don't like having a small > overxxx segment for every datxxx segment that memresize creates.. By > creating the new file my self, I don't have a lot of these small overxxx > segments that are never used. > > I also wrote a process to Select the old file and create a SAVELIST. > Without going into to much detail, the process uses PHANTOM to spawn off > a number of Unidata copies. So there are a number of simultaneous > processes working to build the new file. Each PHANTOM Copy knows how > many total phantoms are working on the file and what sequence IT is > within the total number of PHANTOMS. Each Phantom handles part of the > SAVELIST. Each PHANTOM does the iteration below until the list is > exhausted: > > 1. QSELECT SAVEDLISTS listname000. (process 2 would use 001, process 3 > would use 002 etc) > 2. COPY FROM old.file TO new.file > 3. Increment the list counter from 000 by the number of Phantoms and go > back to step one to process a new segment of the savelist. > 4. If a process cannot find the next savelist segment in SAVEDLISTS, it > is done. > > This process is almost as fast as memresize. You can control what file > systems are used for space reasons and you don't get scads of overxxx > segments in your FILE.NAME directory (one of my dynamic files has 39 dat > segments and just a over001). > > So I don't use memresize for dynamic files anymore. > > -Original Message- > From: u2-users-boun...@listserver.u2ug.org > [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Jonathan > Leckie > Sent: Wednesday, April 25, 2012 9:36 AM > To: u2-users@listserver.u2ug.org > Subject: [U2] Huge Dynamic Unidata file > > I have a very large file that I don't have enough free space to > memresize, however howabout I create new dynamic (temporary) file and > then copy all the records (in ECL) to the new file and then (unix) copy > the temporary file over the top of the original. > > Does that seem like a sensible idea? > > > Regards > Jonathan Leckie > > > > > ** > * This message has been scanned for viruses and dangerous content > * and is believed to be clean. > * > > * This email and any files transmitted with it are confidential and > * intended solely for the use of the individual or entity to whom they > * are addressed. > * > * If you have received this email in error please notify us at Blairs, > * details can be found on our website http://www.blairswindows.co.uk > * > * Name & Registered Of
Re: [U2] Huge Dynamic Unidata file
Yes, absolutely - I'm a great believer in a minimum modulo. If I have to copy large files from one to another my preferred method is to drive a load of PHANTOMs with save-lists in parallel. Also useful to drive from a SELECT not a SSELECT so you get a file-sequential order by "chunking" the input list as well so each PHANTOM gets a file-sequential ordered set of input data to process. Large disk caches can futz this idea a little, but it still holds true in general terms. I think someone else mentioned this as well recently. Regards JayJay On 25 Apr 2012, at 21:28, Wols Lists wrote: > On 25/04/12 15:46, John Jenkins wrote: >> We've recently added a new UniData tuneable to udtconfig "UDT_SPLIT_POLICY" >> which can help conserve space when an overflowed dynamic file splits. The >> total size of the contents are not necessarily the same as the physical file >> size. Always worth checking with "guide" and the latest fixes and changes >> for this change if you have a stake here as the new split policy needs to be >> positively chosen. >> >> Regards >> >> JayJay >> > Dunno Unidata, but if you're copying to a dynamic file, would it make > sense to use MINIMUM_MODULUS on the new file? JayJay, you'd know far > more than me about this, but in UniVerse, it makes sense to use it if > your file is not going to shrink and you're creating it specifically to > populate it with a large amount of pre-existing data. > > Cheers, > Wol > > ___ > U2-Users mailing list > U2-Users@listserver.u2ug.org > http://listserver.u2ug.org/mailman/listinfo/u2-users ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Huge Dynamic Unidata file
I think at the time I wrote this, Unidata 6 I think, I could not even get memresize to handle overflow that was larger than 2 gigabytes. So I thought the solution by Unidata was to create an overflow file for every dat segment. I have never specified the OVERFLOW option when using memresize. Plus, I rarely have enough space in a file system to handle 2 copies of one of my big file. One work around for the file system space issue is to symbolically link the original file in another file system. But that is a lot of work when you have a lot of dat files. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of dean.armbrus...@ferguson.com Sent: Wednesday, April 25, 2012 3:16 PM To: u2-users@listserver.u2ug.org; jonathan.lec...@blairswindows.co.uk Subject: Re: [U2] Huge Dynamic Unidata file Were you using the OVERFLOW option with memresize? If not, memresize should not be creating the extra over files. If memresize did create the extra over files without the OVERFLOW option, then that would be a bug in memresize. Dean Armbruster System Analyst 757-989-2839 -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: Wednesday, April 25, 2012 12:29 To: 'Jonathan Leckie'; 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file I resize most of my Dynamic files this way. I don't like having a small overxxx segment for every datxxx segment that memresize creates.. By creating the new file my self, I don't have a lot of these small overxxx segments that are never used. I also wrote a process to Select the old file and create a SAVELIST. Without going into to much detail, the process uses PHANTOM to spawn off a number of Unidata copies. So there are a number of simultaneous processes working to build the new file. Each PHANTOM Copy knows how many total phantoms are working on the file and what sequence IT is within the total number of PHANTOMS. Each Phantom handles part of the SAVELIST. Each PHANTOM does the iteration below until the list is exhausted: 1. QSELECT SAVEDLISTS listname000. (process 2 would use 001, process 3 would use 002 etc) 2. COPY FROM old.file TO new.file 3. Increment the list counter from 000 by the number of Phantoms and go back to step one to process a new segment of the savelist. 4. If a process cannot find the next savelist segment in SAVEDLISTS, it is done. This process is almost as fast as memresize. You can control what file systems are used for space reasons and you don't get scads of overxxx segments in your FILE.NAME directory (one of my dynamic files has 39 dat segments and just a over001). So I don't use memresize for dynamic files anymore. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Jonathan Leckie Sent: Wednesday, April 25, 2012 9:36 AM To: u2-users@listserver.u2ug.org Subject: [U2] Huge Dynamic Unidata file I have a very large file that I don't have enough free space to memresize, however howabout I create new dynamic (temporary) file and then copy all the records (in ECL) to the new file and then (unix) copy the temporary file over the top of the original. Does that seem like a sensible idea? Regards Jonathan Leckie ** * This message has been scanned for viruses and dangerous content * and is believed to be clean. * * This email and any files transmitted with it are confidential and * intended solely for the use of the individual or entity to whom they * are addressed. * * If you have received this email in error please notify us at Blairs, * details can be found on our website http://www.blairswindows.co.uk * * Name & Registered Office: * * Blairs Windows Limited * Registered office : 9 Baker Street, Greenock, PA15 4TU * Company No: SC393935, V.A.T. registration No: 108729111 ** ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users -- CONFIDENTIALITY NOTICE: If you have received this email in error, please immediately notify the sender by e-mail at the address shown. This email transmission may contain confidential information. This information is intended only for the use of the individual(s) or entity to whom it is intended even if addressed incorrectly. Please delete it from your files if you are not the intended recipient. Thank you for your compliance. Copyright (c) 2012 Cigna == _
Re: [U2] Huge Dynamic Unidata file
On 25/04/12 15:46, John Jenkins wrote: > We've recently added a new UniData tuneable to udtconfig "UDT_SPLIT_POLICY" > which can help conserve space when an overflowed dynamic file splits. The > total size of the contents are not necessarily the same as the physical file > size. Always worth checking with "guide" and the latest fixes and changes for > this change if you have a stake here as the new split policy needs to be > positively chosen. > > Regards > > JayJay > Dunno Unidata, but if you're copying to a dynamic file, would it make sense to use MINIMUM_MODULUS on the new file? JayJay, you'd know far more than me about this, but in UniVerse, it makes sense to use it if your file is not going to shrink and you're creating it specifically to populate it with a large amount of pre-existing data. Cheers, Wol ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Huge Dynamic Unidata file
Were you using the OVERFLOW option with memresize? If not, memresize should not be creating the extra over files. If memresize did create the extra over files without the OVERFLOW option, then that would be a bug in memresize. Dean Armbruster System Analyst 757-989-2839 -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Baakkonen, Rodney A (Rod) 46K Sent: Wednesday, April 25, 2012 12:29 To: 'Jonathan Leckie'; 'U2 Users List' Subject: Re: [U2] Huge Dynamic Unidata file I resize most of my Dynamic files this way. I don't like having a small overxxx segment for every datxxx segment that memresize creates.. By creating the new file my self, I don't have a lot of these small overxxx segments that are never used. I also wrote a process to Select the old file and create a SAVELIST. Without going into to much detail, the process uses PHANTOM to spawn off a number of Unidata copies. So there are a number of simultaneous processes working to build the new file. Each PHANTOM Copy knows how many total phantoms are working on the file and what sequence IT is within the total number of PHANTOMS. Each Phantom handles part of the SAVELIST. Each PHANTOM does the iteration below until the list is exhausted: 1. QSELECT SAVEDLISTS listname000. (process 2 would use 001, process 3 would use 002 etc) 2. COPY FROM old.file TO new.file 3. Increment the list counter from 000 by the number of Phantoms and go back to step one to process a new segment of the savelist. 4. If a process cannot find the next savelist segment in SAVEDLISTS, it is done. This process is almost as fast as memresize. You can control what file systems are used for space reasons and you don't get scads of overxxx segments in your FILE.NAME directory (one of my dynamic files has 39 dat segments and just a over001). So I don't use memresize for dynamic files anymore. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Jonathan Leckie Sent: Wednesday, April 25, 2012 9:36 AM To: u2-users@listserver.u2ug.org Subject: [U2] Huge Dynamic Unidata file I have a very large file that I don't have enough free space to memresize, however howabout I create new dynamic (temporary) file and then copy all the records (in ECL) to the new file and then (unix) copy the temporary file over the top of the original. Does that seem like a sensible idea? Regards Jonathan Leckie ** * This message has been scanned for viruses and dangerous content * and is believed to be clean. * * This email and any files transmitted with it are confidential and * intended solely for the use of the individual or entity to whom they * are addressed. * * If you have received this email in error please notify us at Blairs, * details can be found on our website http://www.blairswindows.co.uk * * Name & Registered Office: * * Blairs Windows Limited * Registered office : 9 Baker Street, Greenock, PA15 4TU * Company No: SC393935, V.A.T. registration No: 108729111 ** ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users -- CONFIDENTIALITY NOTICE: If you have received this email in error, please immediately notify the sender by e-mail at the address shown. This email transmission may contain confidential information. This information is intended only for the use of the individual(s) or entity to whom it is intended even if addressed incorrectly. Please delete it from your files if you are not the intended recipient. Thank you for your compliance. Copyright (c) 2012 Cigna == ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Huge Dynamic Unidata file
I resize most of my Dynamic files this way. I don't like having a small overxxx segment for every datxxx segment that memresize creates.. By creating the new file my self, I don't have a lot of these small overxxx segments that are never used. I also wrote a process to Select the old file and create a SAVELIST. Without going into to much detail, the process uses PHANTOM to spawn off a number of Unidata copies. So there are a number of simultaneous processes working to build the new file. Each PHANTOM Copy knows how many total phantoms are working on the file and what sequence IT is within the total number of PHANTOMS. Each Phantom handles part of the SAVELIST. Each PHANTOM does the iteration below until the list is exhausted: 1. QSELECT SAVEDLISTS listname000. (process 2 would use 001, process 3 would use 002 etc) 2. COPY FROM old.file TO new.file 3. Increment the list counter from 000 by the number of Phantoms and go back to step one to process a new segment of the savelist. 4. If a process cannot find the next savelist segment in SAVEDLISTS, it is done. This process is almost as fast as memresize. You can control what file systems are used for space reasons and you don't get scads of overxxx segments in your FILE.NAME directory (one of my dynamic files has 39 dat segments and just a over001). So I don't use memresize for dynamic files anymore. -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Jonathan Leckie Sent: Wednesday, April 25, 2012 9:36 AM To: u2-users@listserver.u2ug.org Subject: [U2] Huge Dynamic Unidata file I have a very large file that I don't have enough free space to memresize, however howabout I create new dynamic (temporary) file and then copy all the records (in ECL) to the new file and then (unix) copy the temporary file over the top of the original. Does that seem like a sensible idea? Regards Jonathan Leckie ** * This message has been scanned for viruses and dangerous content * and is believed to be clean. * * This email and any files transmitted with it are confidential and * intended solely for the use of the individual or entity to whom they * are addressed. * * If you have received this email in error please notify us at Blairs, * details can be found on our website http://www.blairswindows.co.uk * * Name & Registered Office: * * Blairs Windows Limited * Registered office : 9 Baker Street, Greenock, PA15 4TU * Company No: SC393935, V.A.T. registration No: 108729111 ** ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users -- CONFIDENTIALITY NOTICE: If you have received this email in error, please immediately notify the sender by e-mail at the address shown. This email transmission may contain confidential information. This information is intended only for the use of the individual(s) or entity to whom it is intended even if addressed incorrectly. Please delete it from your files if you are not the intended recipient. Thank you for your compliance. Copyright (c) 2012 Cigna == ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Huge Dynamic Unidata file
Also be sure that anyone that could have access to that file logs out/in before re-accessing the new file. John -Original Message- From: u2-users-boun...@listserver.u2ug.org [mailto:u2-users-boun...@listserver.u2ug.org] On Behalf Of Dave Henderson Sent: Wednesday, April 25, 2012 10:40 AM To: Jonathan Leckie; U2 Users List Cc: u2-users@listserver.u2ug.org Subject: Re: [U2] Huge Dynamic Unidata file Hi, That will work, I have done that many times. Just make sure you check the permissions at unix level after the copy. Dave > I have a very large file that I don't have enough free space to > memresize, however howabout I create new dynamic (temporary) file and > then copy all the records (in ECL) to the new file and then (unix) > copy the temporary file over the top of the original. > > Does that seem like a sensible idea? > > > Regards > Jonathan Leckie > > > > > ** > * This message has been scanned for viruses and dangerous content > * and is believed to be clean. > * > * This email and any files transmitted with it are confidential and > * intended solely for the use of the individual or entity to whom they > * are addressed. > * > * If you have received this email in error please notify us at Blairs, > * details can be found on our website http://www.blairswindows.co.uk > * > * Name & Registered Office: > * > * Blairs Windows Limited > * Registered office : 9 Baker Street, Greenock, PA15 4TU > * Company No: SC393935, V.A.T. registration No: 108729111 > ** > ___ > U2-Users mailing list > U2-Users@listserver.u2ug.org > http://listserver.u2ug.org/mailman/listinfo/u2-users > -- ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Huge Dynamic Unidata file
We've recently added a new UniData tuneable to udtconfig "UDT_SPLIT_POLICY" which can help conserve space when an overflowed dynamic file splits. The total size of the contents are not necessarily the same as the physical file size. Always worth checking with "guide" and the latest fixes and changes for this change if you have a stake here as the new split policy needs to be positively chosen. Regards JayJay On 25 Apr 2012, at 15:39, Dave Henderson wrote: > Hi, > > That will work, I have done that many times. > Just make sure you check the permissions at unix level after the copy. > > Dave > >> I have a very large file that I don't have enough free space to memresize, >> however howabout I create new dynamic (temporary) file and then copy all the >> records (in ECL) to the new file and then (unix) copy the temporary file >> over the top of the original. >> >> Does that seem like a sensible idea? >> >> >> Regards >> Jonathan Leckie >> >> >> >> >> ** >> * This message has been scanned for viruses and dangerous content >> * and is believed to be clean. >> * >> * This email and any files transmitted with it are confidential and >> * intended solely for the use of the individual or entity to whom they >> * are addressed. >> * >> * If you have received this email in error please notify us at Blairs, >> * details can be found on our website http://www.blairswindows.co.uk >> * >> * Name & Registered Office: >> * >> * Blairs Windows Limited >> * Registered office : 9 Baker Street, Greenock, PA15 4TU >> * Company No: SC393935, V.A.T. registration No: 108729111 >> ** >> ___ >> U2-Users mailing list >> U2-Users@listserver.u2ug.org >> http://listserver.u2ug.org/mailman/listinfo/u2-users >> > > > -- > > ___ > U2-Users mailing list > U2-Users@listserver.u2ug.org > http://listserver.u2ug.org/mailman/listinfo/u2-users ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users
Re: [U2] Huge Dynamic Unidata file
Hi, That will work, I have done that many times. Just make sure you check the permissions at unix level after the copy. Dave > I have a very large file that I don't have enough free space to memresize, > however howabout I create new dynamic (temporary) file and then copy all the > records (in ECL) to the new file and then (unix) copy the temporary file > over the top of the original. > > Does that seem like a sensible idea? > > > Regards > Jonathan Leckie > > > > > ** > * This message has been scanned for viruses and dangerous content > * and is believed to be clean. > * > * This email and any files transmitted with it are confidential and > * intended solely for the use of the individual or entity to whom they > * are addressed. > * > * If you have received this email in error please notify us at Blairs, > * details can be found on our website http://www.blairswindows.co.uk > * > * Name & Registered Office: > * > * Blairs Windows Limited > * Registered office : 9 Baker Street, Greenock, PA15 4TU > * Company No: SC393935, V.A.T. registration No: 108729111 > ** > ___ > U2-Users mailing list > U2-Users@listserver.u2ug.org > http://listserver.u2ug.org/mailman/listinfo/u2-users > -- ___ U2-Users mailing list U2-Users@listserver.u2ug.org http://listserver.u2ug.org/mailman/listinfo/u2-users