10 Years of Innovation at IBM

Posted in 2011

Summary

Not a problem thread but a poll: IBM's Jerry Keesee asked the list for the top Informix features and most under-appreciated enhancements since IBM's 2001 acquisition, ahead of his conference keynote. Users nominated large chunks, multiple page/buffer sizes, multiple secondary servers (HDR/RSS/SDS), transaction survival, storage optimization, OAT, auto update statistics, and free developer editions. A side discussion arose on wanting an online "index rebuild"; Art Kagel noted drop/CREATE INDEX ... ONLINE works since v10, but Celso objected that rebuilding PK/FK indexes on billion-row tables still costs days. No resolution to that request is recorded.

Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.

Topics: Server Administration

Jerry Keesee — Tue, 10 May 2011 20:05:48 — source: IIUG Forums & Mailing Lists

July 1st will be the 10th anniversary of the IBM acquisition of the Informix assets. I recently surveyed several of the Informix architects and asked them several questions about the evolution of the Informix technology since the acquisition. I've heard from them. Now I would like to get your thoughts. What do you believe were the top 5 features or enhancements to the Informix portfolio since 2001? What do you believe were the top 3 under appreciated features released over the last 10 years? I appreciate your feedback. I will include some of this feedback in my keynote next week at the Informix conference in Lenexa and then again in the Chat with the Lab on the 18th of May. Regards, Jerry Keesee World Wide Director, Informix Software

↪ replying to Jerry Keesee

Joerg Volz — Tue, 10 May 2011 20:27:24 — source: IIUG Forums & Mailing Lists

Hello Jerry, -bundle IDS 7 with SuSE Linux, that was our re-entry point and future away from SCO and On-Line 5.x. I heard rumors that there are plans to do so with OpenSuSE? I would appreciate it! -very good and detailed documentation -OAT -automatic update statistics -the free developer versions of Informix -at least: not getting Informix buried with DB2 ;-) Continue with your very good job!!! Have fun next week at IIUG Joerg Volz ------------------------------------------------------------------------ ----- -----Original Message----- From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Jerry Keesee Sent: Tuesday, May 10, 2011 10:15 PM To: ids@iiug.org Subject: 10 Years of Innovation at IBM [23627] July 1st will be the 10th anniversary of the IBM acquisition of the Informix assets. I recently surveyed several of the Informix architects and asked them several questions about the evolution of the Informix technology since the acquisition. I've heard from them. Now I would like to get your thoughts. What do you believe were the top 5 features or enhancements to the Informix portfolio since 2001? What do you believe were the top 3 under appreciated features released over the last 10 years? I appreciate your feedback. I will include some of this feedback in my keynote next week at the Informix conference in Lenexa and then again in the Chat with the Lab on the 18th of May. Regards, Jerry Keesee World Wide Director, Informix Software ************************************************************************ ******* Forum Note: Use "Reply" to post a response in the discussion forum. IT Handel und Beratung Jorg Volz Bernhard-Fruh-Str. 7 77855 Achern GERMANY Tel: +49 (0)7841-681651 Fax: +49 (0)7841-681654 Mobil: +49 (0)170-2989757 VAT-ID: DE201383541 http://www.it-volz.de

↪ replying to Jerry Keesee

Art Kagel — Tue, 10 May 2011 20:56:02 — source: IIUG Forums & Mailing Lists

Great idea Jerry! Below: Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. 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 Tue, May 10, 2011 at 4:05 PM, Jerry Keesee <jkeesee@us.ibm.com> wrote: > July 1st will be the 10th anniversary of the IBM acquisition of the > Informix assets. I recently surveyed several of the Informix architects > and asked them several questions about the evolution of the Informix > technology since the acquisition. I've heard from them. Now I would like > to get your thoughts. > > What do you believe were the top 5 features or enhancements to the > Informix portfolio since 2001? > 1. Multiple index query processing. 2. Transaction survival. 3. Expanded storage (large chunks and more chunks/dbspaces supported). 4. Multiple page sizes. 5. Star schema optimization. > What do you believe were the top 3 under appreciated features released > over the last 10 years? > 1. Transaction survival. 2. Multiple secondary servers (HDR, RSS, SDS) > I appreciate your feedback. I will include some of this feedback in my > keynote next week at the Informix conference in Lenexa and then again in > the Chat with the Lab on the 18th of May. > > Regards, > > Jerry Keesee > World Wide Director, Informix Software > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --20cf307cfbca0dff6f04a2f22f7b

↪ replying to Joerg Volz

amarini@fazenda.ms.gov.br — Wed, 11 May 2011 00:33:30 — source: IIUG Forums & Mailing Lists

> Hello Jerry, > > -bundle IDS 7 with SuSE Linux, that was our re-entry point and future > away from SCO and On-Line 5.x. > I heard rumors that there are plans to do so with OpenSuSE? I would > appreciate it! > > -very good and detailed documentation > > -OAT > > -automatic update statistics > > -the free developer versions of Informix > > -at least: not getting Informix buried with DB2 ;-) > > Continue with your very good job!!! > > Have fun next week at IIUG > Joerg Volz > ------------------------------------------------------------------------ > ----- > > -----Original Message----- > From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of > Jerry Keesee > Sent: Tuesday, May 10, 2011 10:15 PM > To: ids@iiug.org > Subject: 10 Years of Innovation at IBM [23627] > > July 1st will be the 10th anniversary of the IBM acquisition of the > Informix assets. I recently surveyed several of the Informix architects > and asked them several questions about the evolution of the Informix > technology since the acquisition. I've heard from them. Now I would like > > to get your thoughts. > > What do you believe were the top 5 features or enhancements to the > Informix portfolio since 2001? > > What do you believe were the top 3 under appreciated features released > over the last 10 years? > > I appreciate your feedback. I will include some of this feedback in my > keynote next week at the Informix conference in Lenexa and then again in > > the Chat with the Lab on the 18th of May. > > Regards, > > Jerry Keesee > World Wide Director, Informix Software > > ************************************************************************ > ******* > Forum Note: Use "Reply" to post a response in the discussion forum. > > IT Handel und Beratung Jorg Volz > Bernhard-Fruh-Str. 7 > 77855 Achern > GERMANY > > Tel: +49 (0)7841-681651 > Fax: +49 (0)7841-681654 > Mobil: +49 (0)170-2989757 > > VAT-ID: DE201383541 > > http://www.it-volz.de > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > >

↪ replying to Art Kagel

rroussey@comcast.net — Wed, 11 May 2011 12:54:31 — source: IIUG Forums & Mailing Lists

- restore with renaming (great for creating dev and QA instances) - multiple secondary server types (greatly expands options for DR) - multiple page sizes (gets you separate buffers / page size) Bob ----- Original Message ----- From: "Art Kagel" <art.kagel@gmail.com> To: ids@iiug.org Sent: Tuesday, May 10, 2011 4:56:03 PM Subject: Re: 10 Years of Innovation at IBM [23630] Great idea Jerry! Below: Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. 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 Tue, May 10, 2011 at 4:05 PM, Jerry Keesee <jkeesee@us.ibm.com> wrote: > July 1st will be the 10th anniversary of the IBM acquisition of the > Informix assets. I recently surveyed several of the Informix architects > and asked them several questions about the evolution of the Informix > technology since the acquisition. I've heard from them. Now I would like > to get your thoughts. > > What do you believe were the top 5 features or enhancements to the > Informix portfolio since 2001? > 1. Multiple index query processing. 2. Transaction survival. 3. Expanded storage (large chunks and more chunks/dbspaces supported). 4. Multiple page sizes. 5. Star schema optimization. > What do you believe were the top 3 under appreciated features released > over the last 10 years? > 1. Transaction survival. 2. Multiple secondary servers (HDR, RSS, SDS) > I appreciate your feedback. I will include some of this feedback in my > keynote next week at the Informix conference in Lenexa and then again in > the Chat with the Lab on the 18th of May. > > Regards, > > Jerry Keesee > World Wide Director, Informix Software > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --20cf307cfbca0dff6f04a2f22f7b ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum. Bob ----- Original Message ----- From: "Art Kagel" <art.kagel@gmail.com> To: ids@iiug.org Sent: Tuesday, May 10, 2011 4:56:03 PM Subject: Re: 10 Years of Innovation at IBM [23630] Great idea Jerry! Below: Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. 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 Tue, May 10, 2011 at 4:05 PM, Jerry Keesee <jkeesee@us.ibm.com> wrote: > July 1st will be the 10th anniversary of the IBM acquisition of the > Informix assets. I recently surveyed several of the Informix architects > and asked them several questions about the evolution of the Informix > technology since the acquisition. I've heard from them. Now I would like > to get your thoughts. > > What do you believe were the top 5 features or enhancements to the > Informix portfolio since 2001? > 1. Multiple index query processing. 2. Transaction survival. 3. Expanded storage (large chunks and more chunks/dbspaces supported). 4. Multiple page sizes. 5. Star schema optimization. > What do you believe were the top 3 under appreciated features released > over the last 10 years? > 1. Transaction survival. 2. Multiple secondary servers (HDR, RSS, SDS) > I appreciate your feedback. I will include some of this feedback in my > keynote next week at the Informix conference in Lenexa and then again in > the Chat with the Lab on the 18th of May. > > Regards, > > Jerry Keesee > World Wide Director, Informix Software > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --20cf307cfbca0dff6f04a2f22f7b ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

↪ replying to Jerry Keesee

Celso Cabral Coimbra — Wed, 11 May 2011 14:18:27 — source: IIUG Forums & Mailing Lists

Jerry, In the sequence 1 - Large chunks 2 - Multiple page size and buffer size 3 - Storage Optimization 4 - Multiple secondary servers 5 - Expand Warehouse Infrastructure 6 - Reduced Overhead For Foreign Key Constraints BTW, I would like to see "index rebuild" (online) in the next version! Best Regards, Celso Cabral Coimbra Administrador de Banco de Dados ClearTech Ltda "Trust at the heart of Communications" Tel. (11) 3576-4509 -----Mensagem original----- De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Jerry Keesee Enviada em: terça-feira, 10 de maio de 2011 17:06 Para: ids@iiug.org Assunto: 10 Years of Innovation at IBM [23627] July 1st will be the 10th anniversary of the IBM acquisition of the Informix assets. I recently surveyed several of the Informix architects and asked them several questions about the evolution of the Informix technology since the acquisition. I've heard from them. Now I would like to get your thoughts. What do you believe were the top 5 features or enhancements to the Informix portfolio since 2001? What do you believe were the top 3 under appreciated features released over the last 10 years? I appreciate your feedback. I will include some of this feedback in my keynote next week at the Informix conference in Lenexa and then again in the Chat with the Lab on the 18th of May. Regards, Jerry Keesee World Wide Director, Informix Software ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

↪ replying to Celso Cabral Coimbra

Art Kagel — Wed, 11 May 2011 14:36:31 — source: IIUG Forums & Mailing Lists

Celso: You can drop and index and build the index online since version 10.00 by adding the ONLINE keyword: begin work; drop index fred_pk; create index fred_pk on fred( serial_col ) online;commit work; Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. 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 Wed, May 11, 2011 at 10:18 AM, Celso Cabral Coimbra < ccoimbra@cleartech.com.br> wrote: > Jerry, > > In the sequence > 1 - Large chunks > 2 - Multiple page size and buffer size > 3 - Storage Optimization > 4 - Multiple secondary servers > 5 - Expand Warehouse Infrastructure > 6 - Reduced Overhead For Foreign Key Constraints > > BTW, I would like to see "index rebuild" (online) in the next version! > > Best Regards, > > Celso Cabral Coimbra > Administrador de Banco de Dados > ClearTech Ltda > "Trust at the heart of Communications" > Tel. (11) 3576-4509 > > -----Mensagem original----- > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Jerry > Keesee > Enviada em: terça-feira, 10 de maio de 2011 17:06 > Para: ids@iiug.org > Assunto: 10 Years of Innovation at IBM [23627] > > July 1st will be the 10th anniversary of the IBM acquisition of the > Informix assets. I recently surveyed several of the Informix architects > and asked them several questions about the evolution of the Informix > technology since the acquisition. I've heard from them. Now I would like > to get your thoughts. > > What do you believe were the top 5 features or enhancements to the > Informix portfolio since 2001? > > What do you believe were the top 3 under appreciated features released > over the last 10 years? > > I appreciate your feedback. I will include some of this feedback in my > keynote next week at the Informix conference in Lenexa and then again in > the Chat with the Lab on the 18th of May. > > Regards, > > Jerry Keesee > World Wide Director, Informix Software > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --20cf3071c6d2861a6b04a300ff82

↪ replying to Art Kagel

Celso Cabral Coimbra — Wed, 11 May 2011 15:01:28 — source: IIUG Forums & Mailing Lists

Yes, I can do that. But imagine dropping a index in a table with 5 bi rows that support a Foreign key to a table with 2 bi rows. Using a supposed "rebuild index online feature", it would recreate a index reading directly the existent index create the new index using other disk space and the drop the old index. The constraints will always have the index to be supported. When I drop a index that support a PK or FK it will create another index automatically to support those constraint, if I'm not mistaken. Thank you Art, Celso Cabral Coimbra Administrador de Banco de Dados ClearTech Ltda "Trust at the heart of Communications" Tel. (11) 3576-4509 -----Mensagem original----- De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Art Kagel Enviada em: quarta-feira, 11 de maio de 2011 11:37 Para: ids@iiug.org Assunto: Re: 10 Years of Innovation at IBM [23644] Celso: You can drop and index and build the index online since version 10.00 by adding the ONLINE keyword: begin work; drop index fred_pk; create index fred_pk on fred( serial_col ) online;commit work; Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. 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 Wed, May 11, 2011 at 10:18 AM, Celso Cabral Coimbra < ccoimbra@cleartech.com.br> wrote: > Jerry, > > In the sequence > 1 - Large chunks > 2 - Multiple page size and buffer size > 3 - Storage Optimization > 4 - Multiple secondary servers > 5 - Expand Warehouse Infrastructure > 6 - Reduced Overhead For Foreign Key Constraints > > BTW, I would like to see "index rebuild" (online) in the next version! > > Best Regards, > > Celso Cabral Coimbra > Administrador de Banco de Dados > ClearTech Ltda > "Trust at the heart of Communications" > Tel. (11) 3576-4509 > > -----Mensagem original----- > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Jerry > Keesee > Enviada em: terça-feira, 10 de maio de 2011 17:06 > Para: ids@iiug.org > Assunto: 10 Years of Innovation at IBM [23627] > > July 1st will be the 10th anniversary of the IBM acquisition of the > Informix assets. I recently surveyed several of the Informix architects > and asked them several questions about the evolution of the Informix > technology since the acquisition. I've heard from them. Now I would like > to get your thoughts. > > What do you believe were the top 5 features or enhancements to the > Informix portfolio since 2001? > > What do you believe were the top 3 under appreciated features released > over the last 10 years? > > I appreciate your feedback. I will include some of this feedback in my > keynote next week at the Informix conference in Lenexa and then again in > the Chat with the Lab on the 18th of May. > > Regards, > > Jerry Keesee > World Wide Director, Informix Software > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --20cf3071c6d2861a6b04a300ff82 ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

↪ replying to Jerry Keesee

Peter_Logan@spartanstores.com — Wed, 11 May 2011 15:06:49 — source: IIUG Forums & Mailing Lists

Jerry, In no particular order: Storage optimization Multiple page/buffer sizes Single user mode Online index create/drop Large chunk support Peter Logan Senior Database Administrator Phone: 616/878-8309 From: "Jerry Keesee" <jkeesee@us.ibm.com> To: ids@iiug.org Date: 05/10/2011 04:06 PM Subject: 10 Years of Innovation at IBM [23627] Sent by: ids-bounces@iiug.org July 1st will be the 10th anniversary of the IBM acquisition of the Informix assets. I recently surveyed several of the Informix architects and asked them several questions about the evolution of the Informix technology since the acquisition. I've heard from them. Now I would like to get your thoughts. What do you believe were the top 5 features or enhancements to the Informix portfolio since 2001? What do you believe were the top 3 under appreciated features released over the last 10 years? I appreciate your feedback. I will include some of this feedback in my keynote next week at the Informix conference in Lenexa and then again in the Chat with the Lab on the 18th of May. Regards, Jerry Keesee World Wide Director, Informix Software ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

↪ replying to Celso Cabral Coimbra

Fernando Nunes — Wed, 11 May 2011 15:07:41 — source: IIUG Forums & Mailing Lists

Not sure if I'm following this... Why would we drop and index and create it again? One situation comes to mind that is, if we want to "clean" the index. For that, the "rebuild" would be potencially helpful since we could skip the ORDER of the records. Is this what you're talking about? Also, sometimes exclsuive locks could potencially be exchanges bey shared locks (no changes) Regards. On Wed, May 11, 2011 at 4:01 PM, Celso Cabral Coimbra < ccoimbra@cleartech.com.br> wrote: > Yes, I can do that. But imagine dropping a index in a table with 5 bi rows > that support a Foreign key to a table with 2 bi rows. > Using a supposed "rebuild index online feature", it would recreate a index > reading directly the existent index create the new index using other disk > space and the drop the old index. The constraints will always have the > index > to be supported. > When I drop a index that support a PK or FK it will create another index > automatically to support those constraint, if I'm not mistaken. > > Thank you Art, > > Celso Cabral Coimbra > Administrador de Banco de Dados > ClearTech Ltda > "Trust at the heart of Communications" > Tel. (11) 3576-4509 > > -----Mensagem original----- > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Art > Kagel > Enviada em: quarta-feira, 11 de maio de 2011 11:37 > Para: ids@iiug.org > Assunto: Re: 10 Years of Innovation at IBM [23644] > > Celso: > > You can drop and index and build the index online since version 10.00 by > adding the ONLINE keyword: > > begin work; > drop index fred_pk; > create index fred_pk on fred( serial_col ) online;> commit work; > > Art > > Art S. Kagel > Advanced DataTools (www.advancedatatools.com) > Blog: http://informix-myview.blogspot.com/ > > Disclaimer: Please keep in mind that my own opinions are my own opinions > and > do not reflect on my employer, Advanced DataTools, the IIUG, nor any other > organization with which I am associated either explicitly, implicitly, or > by > inference. 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 Wed, May 11, 2011 at 10:18 AM, Celso Cabral Coimbra < > ccoimbra@cleartech.com.br> wrote: > > > Jerry, > > > > In the sequence > > 1 - Large chunks > > 2 - Multiple page size and buffer size > > 3 - Storage Optimization > > 4 - Multiple secondary servers > > 5 - Expand Warehouse Infrastructure > > 6 - Reduced Overhead For Foreign Key Constraints > > > > BTW, I would like to see "index rebuild" (online) in the next version! > > > > Best Regards, > > > > Celso Cabral Coimbra > > Administrador de Banco de Dados > > ClearTech Ltda > > "Trust at the heart of Communications" > > Tel. (11) 3576-4509 > > > > -----Mensagem original----- > > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Jerry > > Keesee > > Enviada em: terça-feira, 10 de maio de 2011 17:06 > > Para: ids@iiug.org > > Assunto: 10 Years of Innovation at IBM [23627] > > > > July 1st will be the 10th anniversary of the IBM acquisition of the > > Informix assets. I recently surveyed several of the Informix architects > > and asked them several questions about the evolution of the Informix > > technology since the acquisition. I've heard from them. Now I would like > > to get your thoughts. > > > > What do you believe were the top 5 features or enhancements to the > > Informix portfolio since 2001? > > > > What do you believe were the top 3 under appreciated features released > > over the last 10 years? > > > > I appreciate your feedback. I will include some of this feedback in my > > keynote next week at the Informix conference in Lenexa and then again in > > the Chat with the Lab on the 18th of May. > > > > Regards, > > > > Jerry Keesee > > World Wide Director, Informix Software > > > > > > > > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > > > > > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > > > --20cf3071c6d2861a6b04a300ff82 > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently... --001485f911f676649104a3016e07

↪ replying to Fernando Nunes

Celso Cabral Coimbra — Wed, 11 May 2011 15:20:21 — source: IIUG Forums & Mailing Lists

Fernando, Yes, it is exactly what I'm talking about. I want to "clean" the index. Dropping and recreating the index, creates it smaller and faster, even creating it using the fillfactor like 70 or smaller. But I have a overhead of days to recreate the PK and mainly FK after recreating the index. This feature exists in others RDBMS and is very helpful. Regards, Celso Cabral Coimbra Administrador de Banco de Dados ClearTech Ltda "Trust at the heart of Communications" Tel. (11) 3576-4509 -----Mensagem original----- De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Fernando Nunes Enviada em: quarta-feira, 11 de maio de 2011 12:08 Para: ids@iiug.org Assunto: Re: 10 Years of Innovation at IBM [23647] Not sure if I'm following this... Why would we drop and index and create it again? One situation comes to mind that is, if we want to "clean" the index. For that, the "rebuild" would be potencially helpful since we could skip the ORDER of the records. Is this what you're talking about? Also, sometimes exclsuive locks could potencially be exchanges bey shared locks (no changes) Regards. On Wed, May 11, 2011 at 4:01 PM, Celso Cabral Coimbra < ccoimbra@cleartech.com.br> wrote: > Yes, I can do that. But imagine dropping a index in a table with 5 bi rows > that support a Foreign key to a table with 2 bi rows. > Using a supposed "rebuild index online feature", it would recreate a index > reading directly the existent index create the new index using other disk > space and the drop the old index. The constraints will always have the > index > to be supported. > When I drop a index that support a PK or FK it will create another index > automatically to support those constraint, if I'm not mistaken. > > Thank you Art, > > Celso Cabral Coimbra > Administrador de Banco de Dados > ClearTech Ltda > "Trust at the heart of Communications" > Tel. (11) 3576-4509 > > -----Mensagem original----- > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Art > Kagel > Enviada em: quarta-feira, 11 de maio de 2011 11:37 > Para: ids@iiug.org > Assunto: Re: 10 Years of Innovation at IBM [23644] > > Celso: > > You can drop and index and build the index online since version 10.00 by > adding the ONLINE keyword: > > begin work; > drop index fred_pk; > create index fred_pk on fred( serial_col ) online;> commit work; > > Art > > Art S. Kagel > Advanced DataTools (www.advancedatatools.com) > Blog: http://informix-myview.blogspot.com/ > > Disclaimer: Please keep in mind that my own opinions are my own opinions > and > do not reflect on my employer, Advanced DataTools, the IIUG, nor any other > organization with which I am associated either explicitly, implicitly, or > by > inference. 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 Wed, May 11, 2011 at 10:18 AM, Celso Cabral Coimbra < > ccoimbra@cleartech.com.br> wrote: > > > Jerry, > > > > In the sequence > > 1 - Large chunks > > 2 - Multiple page size and buffer size > > 3 - Storage Optimization > > 4 - Multiple secondary servers > > 5 - Expand Warehouse Infrastructure > > 6 - Reduced Overhead For Foreign Key Constraints > > > > BTW, I would like to see "index rebuild" (online) in the next version! > > > > Best Regards, > > > > Celso Cabral Coimbra > > Administrador de Banco de Dados > > ClearTech Ltda > > "Trust at the heart of Communications" > > Tel. (11) 3576-4509 > > > > -----Mensagem original----- > > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Jerry > > Keesee > > Enviada em: terça-feira, 10 de maio de 2011 17:06 > > Para: ids@iiug.org > > Assunto: 10 Years of Innovation at IBM [23627] > > > > July 1st will be the 10th anniversary of the IBM acquisition of the > > Informix assets. I recently surveyed several of the Informix architects > > and asked them several questions about the evolution of the Informix > > technology since the acquisition. I've heard from them. Now I would like > > to get your thoughts. > > > > What do you believe were the top 5 features or enhancements to the > > Informix portfolio since 2001? > > > > What do you believe were the top 3 under appreciated features released > > over the last 10 years? > > > > I appreciate your feedback. I will include some of this feedback in my > > keynote next week at the Informix conference in Lenexa and then again in > > the Chat with the Lab on the 18th of May. > > > > Regards, > > > > Jerry Keesee > > World Wide Director, Informix Software > > > > > > > > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > > > > > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > > > --20cf3071c6d2861a6b04a300ff82 > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently... --001485f911f676649104a3016e07 ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

↪ replying to Celso Cabral Coimbra

Art Kagel — Wed, 11 May 2011 15:31:04 — source: IIUG Forums & Mailing Lists

Celso: You have a valid point about constraint indexes. Actually, if you drop the independent index that supports a constraint, it doesn't rebuild the index - that would be good for your purposes - it just renames the existing index to a hidden name and patches up the constraint's link to the index. Hmm, a way to easily perform an online reorg of an index will have some uses. As Celso points out in his response to Fernando, after some operations, like massive inserts or deletes, the index will become inefficient with many half full pages or pages with several deleted keys on them. Mostly the btree scanner threads will eventually clean and compress the index that had deletes, but only over time and only for those pages that have recently been accessed after the deletes due to the way that ALICE mode works. The index suffering from many half full pages due to massive numbers of inserts will not be cleaned and compressed in the same way at all though. Many queries will be less efficient than they could be if the index were to be reorged with a high FILLFACTOR - at least for a while. Jerry, add this one to my list as well - a way to reorg an index online, including constraint indexes, with an optional FILLFACTOR! Celso is on to something. Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. 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 Wed, May 11, 2011 at 11:01 AM, Celso Cabral Coimbra < ccoimbra@cleartech.com.br> wrote: > Yes, I can do that. But imagine dropping a index in a table with 5 bi rows > that support a Foreign key to a table with 2 bi rows. > Using a supposed "rebuild index online feature", it would recreate a index > reading directly the existent index create the new index using other disk > space and the drop the old index. The constraints will always have the > index > to be supported. > When I drop a index that support a PK or FK it will create another index > automatically to support those constraint, if I'm not mistaken. > > Thank you Art, > > Celso Cabral Coimbra > Administrador de Banco de Dados > ClearTech Ltda > "Trust at the heart of Communications" > Tel. (11) 3576-4509 > > -----Mensagem original----- > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Art > Kagel > Enviada em: quarta-feira, 11 de maio de 2011 11:37 > Para: ids@iiug.org > Assunto: Re: 10 Years of Innovation at IBM [23644] > > Celso: > > You can drop and index and build the index online since version 10.00 by > adding the ONLINE keyword: > > begin work; > drop index fred_pk; > create index fred_pk on fred( serial_col ) online;> commit work; > > Art > > Art S. Kagel > Advanced DataTools (www.advancedatatools.com) > Blog: http://informix-myview.blogspot.com/ > > Disclaimer: Please keep in mind that my own opinions are my own opinions > and > do not reflect on my employer, Advanced DataTools, the IIUG, nor any other > organization with which I am associated either explicitly, implicitly, or > by > inference. 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 Wed, May 11, 2011 at 10:18 AM, Celso Cabral Coimbra < > ccoimbra@cleartech.com.br> wrote: > > > Jerry, > > > > In the sequence > > 1 - Large chunks > > 2 - Multiple page size and buffer size > > 3 - Storage Optimization > > 4 - Multiple secondary servers > > 5 - Expand Warehouse Infrastructure > > 6 - Reduced Overhead For Foreign Key Constraints > > > > BTW, I would like to see "index rebuild" (online) in the next version! > > > > Best Regards, > > > > Celso Cabral Coimbra > > Administrador de Banco de Dados > > ClearTech Ltda > > "Trust at the heart of Communications" > > Tel. (11) 3576-4509 > > > > -----Mensagem original----- > > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Jerry > > Keesee > > Enviada em: terça-feira, 10 de maio de 2011 17:06 > > Para: ids@iiug.org > > Assunto: 10 Years of Innovation at IBM [23627] > > > > July 1st will be the 10th anniversary of the IBM acquisition of the > > Informix assets. I recently surveyed several of the Informix architects > > and asked them several questions about the evolution of the Informix > > technology since the acquisition. I've heard from them. Now I would like > > to get your thoughts. > > > > What do you believe were the top 5 features or enhancements to the > > Informix portfolio since 2001? > > > > What do you believe were the top 3 under appreciated features released > > over the last 10 years? > > > > I appreciate your feedback. I will include some of this feedback in my > > keynote next week at the Informix conference in Lenexa and then again in > > the Chat with the Lab on the 18th of May. > > > > Regards, > > > > Jerry Keesee > > World Wide Director, Informix Software > > > > > > > > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > > > > > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > > > --20cf3071c6d2861a6b04a300ff82 > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --20cf3071d02e6d0fbf04a301c2ed

↪ replying to Celso Cabral Coimbra

John Miller iii — Wed, 11 May 2011 15:43:25 — source: IIUG Forums & Mailing Lists

If your ultimate goal is to "clean" the index then it sounds like you should ensure the btree scanners are properly tuned and then request that they= clean all of your indexes. This will compact your indexes online wit= hout having to rebuild the index. While this functionality of cleaning the indexes has been in for quite some time, there were big improvement to this cleaning functionality in version 11.50.xC2. If you want more details, I will be talking about this in my performan= ce tutorial at the IIUG conference this Sunday. -:) John F. Miller III STSM, Embedability Architect miller3@us.ibm.com 503-578-5645 IBM Informix Dynamic Server (IDS) ids-bounces@iiug.org wrote on 05/11/2011 08:20:21 AM: > From: > > "Celso Cabral Coimbra" <ccoimbra@cleartech.com.br> > > To: > > ids@iiug.org > > Date: > > 05/11/2011 08:21 AM > > Subject: > > RES: 10 Years of Innovation at IBM [23648] > > Sent by: > > ids-bounces@iiug.org > > Fernando, > > Yes, it is exactly what I'm talking about. I want to "clean" the inde= x. > Dropping and recreating the index, creates it smaller and faster, eve= n > creating it using the fillfactor like 70 or smaller. But I have a overhead of > days to recreate the PK and mainly FK after recreating the index. > This feature exists in others RDBMS and is very helpful. > > Regards, > > Celso Cabral Coimbra > Administrador de Banco de Dados > ClearTech Ltda > "Trust at the heart of Communications" > Tel. (11) 3576-4509 > > -----Mensagem original----- > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de Fernando > Nunes > Enviada em: quarta-feira, 11 de maio de 2011 12:08 > Para: ids@iiug.org > Assunto: Re: 10 Years of Innovation at IBM [23647] > > Not sure if I'm following this... Why would we drop and index and cre= ate it > again? > One situation comes to mind that is, if we want to "clean" the index.= For > that, the "rebuild" would be potencially helpful since we could skip = the > ORDER of the records. > Is this what you're talking about? > > Also, sometimes exclsuive locks could potencially be exchanges bey sh= ared > locks (no changes) > Regards. > > On Wed, May 11, 2011 at 4:01 PM, Celso Cabral Coimbra < > ccoimbra@cleartech.com.br> wrote: > > > Yes, I can do that. But imagine dropping a index in a table with 5 = bi rows > > that support a Foreign key to a table with 2 bi rows. > > Using a supposed "rebuild index online feature", it would recreate = a index > > reading directly the existent index create the new index using othe= r disk > > space and the drop the old index. The constraints will always have = the > > index > > to be supported. > > When I drop a index that support a PK or FK it will create another index > > automatically to support those constraint, if I'm not mistaken. > > > > Thank you Art, > > > > Celso Cabral Coimbra > > Administrador de Banco de Dados > > ClearTech Ltda > > "Trust at the heart of Communications" > > Tel. (11) 3576-4509 > > > > -----Mensagem original----- > > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de A= rt > > Kagel > > Enviada em: quarta-feira, 11 de maio de 2011 11:37 > > Para: ids@iiug.org > > Assunto: Re: 10 Years of Innovation at IBM [23644] > > > > Celso: > > > > You can drop and index and build the index online since version 10.= 00 by > > adding the ONLINE keyword: > > > > begin work; > > drop index fred_pk; > > create index fred_pk on fred( serial_col ) online;> > commit work; > > > > Art > > > > Art S. Kagel > > Advanced DataTools (www.advancedatatools.com) > > Blog: http://informix-myview.blogspot.com/ > > > > Disclaimer: Please keep in mind that my own opinions are my own opinions > > and > > do not reflect on my employer, Advanced DataTools, the IIUG, nor an= y other > > organization with which I am associated either explicitly, implicit= ly, or > > by > > inference. Neither do those opinions reflect those of other individ= uals > > affiliated with any entity with which I am affiliated nor those of = the > > entities themselves. > > > > On Wed, May 11, 2011 at 10:18 AM, Celso Cabral Coimbra < > > ccoimbra@cleartech.com.br> wrote: > > > > > Jerry, > > > > > > In the sequence > > > 1 - Large chunks > > > 2 - Multiple page size and buffer size > > > 3 - Storage Optimization > > > 4 - Multiple secondary servers > > > 5 - Expand Warehouse Infrastructure > > > 6 - Reduced Overhead For Foreign Key Constraints > > > > > > BTW, I would like to see "index rebuild" (online) in the next version! > > > > > > Best Regards, > > > > > > Celso Cabral Coimbra > > > Administrador de Banco de Dados > > > ClearTech Ltda > > > "Trust at the heart of Communications" > > > Tel. (11) 3576-4509 > > > > > > -----Mensagem original----- > > > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de= Jerry > > > Keesee > > > Enviada em: ter=E7a-feira, 10 de maio de 2011 17:06 > > > Para: ids@iiug.org > > > Assunto: 10 Years of Innovation at IBM [23627] > > > > > > July 1st will be the 10th anniversary of the IBM acquisition of t= he > > > Informix assets. I recently surveyed several of the Informix architects > > > and asked them several questions about the evolution of the Infor= mix > > > technology since the acquisition. I've heard from them. Now I wou= ld like > > > to get your thoughts. > > > > > > What do you believe were the top 5 features or enhancements to th= e > > > Informix portfolio since 2001? > > > > > > What do you believe were the top 3 under appreciated features released > > > over the last 10 years? > > > > > > I appreciate your feedback. I will include some of this feedback = in my > > > keynote next week at the Informix conference in Lenexa and then a= gain in > > > the Chat with the Lab on the 18th of May. > > > > > > Regards, > > > > > > Jerry Keesee > > > World Wide Director, Informix Software > > > > > > > > > > > > > > > > > > > > ***********************************************************************= ******** > > > Forum Note: Use "Reply" to post a response in the discussion foru= m. > > > > > > > > > > > > > > > > > > > > ***********************************************************************= ******** > > > Forum Note: Use "Reply" to post a response in the discussion foru= m. > > > > > > > > > > --20cf3071c6d2861a6b04a300ff82 > > > > > > > > > > ***********************************************************************= ******** > > Forum Note: Use "Reply" to post a response in the discussion forum.=@

↪ replying to John Miller iii

Fernando Nunes — Wed, 11 May 2011 15:55:02 — source: IIUG Forums & Mailing Lists

I think that the btree cleaners can't "clean" the indexes properly in the sense I'm thinking about. This of course is not an expertise opinion, but the result of some past experiences. My feeling is that if your index has some special characteristics it will become worst with time. Specifically I'm thinking in indexes containing timestamps. The insert order will mean that new records will have keys "bigger" than all the others... I don't know how to explain this technically, but you'll see your performance suffer in some particular situations (update stats low is the most obvious). And yes, I know this particular problem is being analyzed/fixed, and I'm aware that it's not trivial (clustered value for the optimizer). But the same causes can also cause performance degradation in queries that retrieve large number of keys from the index (key first). Although SKIP SCAN should solve this... Now... If you have some custom magic that will make the btree scaners solve this issue I'm all hears... unfortunately I won't be at the conference... :( Regards On Wed, May 11, 2011 at 4:43 PM, John Miller iii <miller3@us.ibm.com> wrote: > If your ultimate goal is to "clean" the index then it sounds like you > should > ensure the btree scanners are properly tuned and then request that they= > > clean all of your indexes. This will compact your indexes online wit= > hout > having to rebuild the index. While this functionality of cleaning the > indexes > has been in for quite some time, there were big improvement to this > cleaning functionality in version 11.50.xC2. > > If you want more details, I will be talking about this in my performan= > ce > tutorial at the IIUG conference this Sunday. -:) > > John F. Miller III > STSM, Embedability Architect > miller3@us.ibm.com > 503-578-5645 > IBM Informix Dynamic Server (IDS) > > ids-bounces@iiug.org wrote on 05/11/2011 08:20:21 AM: > > > From: > > > > "Celso Cabral Coimbra" <ccoimbra@cleartech.com.br> > > > > To: > > > > ids@iiug.org > > > > Date: > > > > 05/11/2011 08:21 AM > > > > Subject: > > > > RES: 10 Years of Innovation at IBM [23648] > > > > Sent by: > > > > ids-bounces@iiug.org > > > > Fernando, > > > > Yes, it is exactly what I'm talking about. I want to "clean" the inde= > x. > > Dropping and recreating the index, creates it smaller and faster, eve= > n > > creating it using the fillfactor like 70 or smaller. But I have a > overhead of > > days to recreate the PK and mainly FK after recreating the index. > > This feature exists in others RDBMS and is very helpful. > > > > Regards, > > > > Celso Cabral Coimbra > > Administrador de Banco de Dados > > ClearTech Ltda > > "Trust at the heart of Communications" > > Tel. (11) 3576-4509 > > > > -----Mensagem original----- > > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de > Fernando > > Nunes > > Enviada em: quarta-feira, 11 de maio de 2011 12:08 > > Para: ids@iiug.org > > Assunto: Re: 10 Years of Innovation at IBM [23647] > > > > Not sure if I'm following this... Why would we drop and index and cre= > ate > it > > again? > > One situation comes to mind that is, if we want to "clean" the index.= > For > > > that, the "rebuild" would be potencially helpful since we could skip = > the > > ORDER of the records. > > Is this what you're talking about? > > > > Also, sometimes exclsuive locks could potencially be exchanges bey sh= > ared > > > locks (no changes) > > Regards. > > > > On Wed, May 11, 2011 at 4:01 PM, Celso Cabral Coimbra < > > ccoimbra@cleartech.com.br> wrote: > > > > > Yes, I can do that. But imagine dropping a index in a table with 5 = > bi > rows > > > that support a Foreign key to a table with 2 bi rows. > > > Using a supposed "rebuild index online feature", it would recreate = > a > index > > > reading directly the existent index create the new index using othe= > r > disk > > > space and the drop the old index. The constraints will always have = > the > > > index > > > to be supported. > > > When I drop a index that support a PK or FK it will create another > index > > > automatically to support those constraint, if I'm not mistaken. > > > > > > Thank you Art, > > > > > > Celso Cabral Coimbra > > > Administrador de Banco de Dados > > > ClearTech Ltda > > > "Trust at the heart of Communications" > > > Tel. (11) 3576-4509 > > > > > > -----Mensagem original----- > > > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de A= > rt > > > Kagel > > > Enviada em: quarta-feira, 11 de maio de 2011 11:37 > > > Para: ids@iiug.org > > > Assunto: Re: 10 Years of Innovation at IBM [23644] > > > > > > Celso: > > > > > > You can drop and index and build the index online since version 10.= > 00 > by > > > adding the ONLINE keyword: > > > > > > begin work; > > > drop index fred_pk; > > > create index fred_pk on fred( serial_col ) online;> > > commit work; > > > > > > Art > > > > > > Art S. Kagel > > > Advanced DataTools (www.advancedatatools.com) > > > Blog: http://informix-myview.blogspot.com/ > > > > > > Disclaimer: Please keep in mind that my own opinions are my own > opinions > > > and > > > do not reflect on my employer, Advanced DataTools, the IIUG, nor an= > y > other > > > organization with which I am associated either explicitly, implicit= > ly, > or > > > by > > > inference. Neither do those opinions reflect those of other individ= > uals > > > > affiliated with any entity with which I am affiliated nor those of = > the > > > entities themselves. > > > > > > On Wed, May 11, 2011 at 10:18 AM, Celso Cabral Coimbra < > > > ccoimbra@cleartech.com.br> wrote: > > > > > > > Jerry, > > > > > > > > In the sequence > > > > 1 - Large chunks > > > > 2 - Multiple page size and buffer size > > > > 3 - Storage Optimization > > > > 4 - Multiple secondary servers > > > > 5 - Expand Warehouse Infrastructure > > > > 6 - Reduced Overhead For Foreign Key Constraints > > > > > > > > BTW, I would like to see "index rebuild" (online) in the next > version! > > > > > > > > Best Regards, > > > > > > > > Celso Cabral Coimbra > > > > Administrador de Banco de Dados > > > > ClearTech Ltda > > > > "Trust at the heart of Communications" > > > > Tel. (11) 3576-4509 > > > > > > > > -----Mensagem original----- > > > > De: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] Em nome de= > > Jerry > > > > Keesee > > > > Enviada em: ter=E7a-feira, 10 de maio de 2011 17:06 > > > > Para: ids@iiug.org > > > > Assunto: 10 Years of Innovation at IBM [23627] > > > > > > > > July 1st will be the 10th anniversary of the IBM acquisition of t= > he > > > > Informix assets. I recently surveyed several of the Informix > architects > > > > and asked them several questions about th

↪ replying to Jerry Keesee

Fernando Nunes — Wed, 11 May 2011 21:07:53 — source: IIUG Forums & Mailing Lists

On Tue, May 10, 2011 at 9:05 PM, Jerry Keesee <jkeesee@us.ibm.com> wrote: > July 1st will be the 10th anniversary of the IBM acquisition of the > Informix assets. I recently surveyed several of the Informix architects > and asked them several questions about the evolution of the Informix > technology since the acquisition. I've heard from them. Now I would like > to get your thoughts. > > What do you believe were the top 5 features or enhancements to the > Informix portfolio since 2001? > - MACH 11 (is it fair to use just one "feature" to include all of it? - DRDA/Common Client - LAST COMMITTED READ - SQL Admin API / OAT - Large Chunks But it's impossible to say. For each one I choose I can think of a specific customer that would choose another... > > What do you believe were the top 3 under appreciated features released > over the last 10 years? > - DRDA / Common Client - LAST COMMITTED READ - SQL Admin API / OAT > > I appreciate your feedback. I will include some of this feedback in my > keynote next week at the Informix conference in Lenexa and then again in > the Chat with the Lab on the 18th of May. > Regards. -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently... --0016361648ad172be304a3067713

↪ replying to Jerry Keesee

Obnoxio The Clown — Wed, 11 May 2011 21:50:02 — source: IIUG Forums & Mailing Lists

On 10/05/2011 21:05, Jerry Keesee wrote: > What do you believe were the top 3 under appreciated features released > over the last 10 years? Time Series. :o) -- Cheers, Obnoxio The Clown http://obotheclown.blogspot.com I will now proceed to pleasure myself with this fish.

↪ replying to Jerry Keesee

Jack Parker — Thu, 12 May 2011 02:37:34 — source: IIUG Forums & Mailing Lists

Jerry, My concern of late is how I remember Informix and how it behaves now. In the olden days Informix was a flexible and pliable beast, there was = nothing you could not do. If something didn't behave the way you needed = it to, there were a dozen ways to work around it to bend it to your = will. What I experience of late is that I need to bend my requirements to how = the software works, Perhaps that's just me, my shoddy memory and my corporate masters. Regards, Jack Parker On May 10, 2011, at 4:05 PM, Jerry Keesee wrote: > July 1st will be the 10th anniversary of the IBM acquisition of the=20 > Informix assets. I recently surveyed several of the Informix = architects=20 > and asked them several questions about the evolution of the Informix=20= > technology since the acquisition. I've heard from them. Now I would = like=20 > to get your thoughts.=20 >=20 > What do you believe were the top 5 features or enhancements to the=20 > Informix portfolio since 2001?=20 >=20 > What do you believe were the top 3 under appreciated features released=20= > over the last 10 years?=20 >=20 > I appreciate your feedback. I will include some of this feedback in my=20= > keynote next week at the Informix conference in Lenexa and then again = in=20 > the Chat with the Lab on the 18th of May.=20 >=20 > Regards,=20 >=20 > Jerry Keesee=20 > World Wide Director, Informix Software=20 >=20 >=20 > = **************************************************************************= *****=20 > Forum Note: Use "Reply" to post a response in the discussion forum.=20= >=20

↪ replying to Jack Parker

Fernando Nunes — Thu, 12 May 2011 10:00:02 — source: IIUG Forums & Mailing Lists

Can you elaborate on this? A few examples? I can think about a lot of things I'd like it to do and it doesn't , but I can't think of one thing it used to do and now doesn't... Regards. On Thu, May 12, 2011 at 3:37 AM, Jack Parker <jack.parker4@verizon.net>wrote: > Jerry, > > My concern of late is how I remember Informix and how it behaves now. > > In the olden days Informix was a flexible and pliable beast, there was = > nothing you could not do. If something didn't behave the way you needed = > it to, there were a dozen ways to work around it to bend it to your = > will. > > What I experience of late is that I need to bend my requirements to how = > the software works, > > Perhaps that's just me, my shoddy memory and my corporate masters. > > Regards, > Jack Parker > > On May 10, 2011, at 4:05 PM, Jerry Keesee wrote: > > > July 1st will be the 10th anniversary of the IBM acquisition of the=20 > > Informix assets. I recently surveyed several of the Informix = > architects=20 > > and asked them several questions about the evolution of the Informix=20= > > > technology since the acquisition. I've heard from them. Now I would = > like=20 > > to get your thoughts.=20 > >=20 > > What do you believe were the top 5 features or enhancements to the=20 > > Informix portfolio since 2001?=20 > >=20 > > What do you believe were the top 3 under appreciated features > released=20= > > > over the last 10 years?=20 > >=20 > > I appreciate your feedback. I will include some of this feedback in > my=20= > > > keynote next week at the Informix conference in Lenexa and then again = > in=20 > > the Chat with the Lab on the 18th of May.=20 > >=20 > > Regards,=20 > >=20 > > Jerry Keesee=20 > > World Wide Director, Informix Software=20 > >=20 > >=20 > > = > **************************************************************************= > *****=20 > > Forum Note: Use "Reply" to post a response in the discussion forum.=20= > > >=20 > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently... --001636c5c22d5e5b7604a31140aa

↪ replying to Fernando Nunes

Jack Parker — Thu, 12 May 2011 11:41:25 — source: IIUG Forums & Mailing Lists

All of the things that you used to be able to do you can still do, it's = the "new" stuff. Try getting HPL to work under windows. There are all sorts of = restrictions. No pipes. No ipload interface, everything has to go = through onpladm, you can't actually invoke the job through onpladm (even = though it is advertised as such - you have to invoke onpload), I forget = now, but I seem to remember issues with device arrays. Work on a mac, no Xtree. I've just spend a couple of days trying to get ifxdeploy to work on Win = 2008. You have to turn your body sideways and approach it in just the = right direction (if you use ifxdeployassist from the command line you = have to be selective in what options you choose (I've given up there and = moved to the GUI until I can get ifxdeploy to work), has to have java = installed and in the PATH, extracts to the local directory instead of = $INFORMIXDIR) - then complains that %INFORMIXDIR/etc/%ONCONFIG doesn't = exist - and currently it fails when trying to allocate dbspaces. I'm = currently working with folks at IBM on this one. Uninstall on Windows is iffy. You have to check that %INFORMIXDIR is = actually gone, %PATH doesn't get cleaned up, services may not have been = cleaned up. I have a list of things I check, and 80% of the time I have = to manually fix them. No fast alter if there's an opaque data type. Building a data management by fragmentation scheme (age off old data by = dropping fragments). You can't have RI turned on, can't have primary = keys, I can't recall all of the things I had to work around on that one, = but it took a couple of months. There is all this "new" stuff, but it no longer seems to be fully = thought through. I love that building an index updates statistics cheers j. On May 12, 2011, at 6:00 AM, Fernando Nunes wrote: > Can you elaborate on this? A few examples?=20 > I can think about a lot of things I'd like it to do and it doesn't , = but I=20 > can't think of one thing it used to do and now doesn't...=20 > Regards.=20 >=20 > On Thu, May 12, 2011 at 3:37 AM, Jack Parker = <jack.parker4@verizon.net>wrote:=20 >=20 >> Jerry,=20 >>=20 >> My concern of late is how I remember Informix and how it behaves now.=20= >>=20 >> In the olden days Informix was a flexible and pliable beast, there = was =3D=20 >> nothing you could not do. If something didn't behave the way you = needed =3D=20 >> it to, there were a dozen ways to work around it to bend it to your =3D= =20 >> will.=20 >>=20 >> What I experience of late is that I need to bend my requirements to = how =3D=20 >> the software works,=20 >>=20 >> Perhaps that's just me, my shoddy memory and my corporate masters.=20 >>=20 >> Regards,=20 >> Jack Parker=20 >>=20 >> On May 10, 2011, at 4:05 PM, Jerry Keesee wrote:=20 >>=20 >>> July 1st will be the 10th anniversary of the IBM acquisition of = the=3D20=20 >>> Informix assets. I recently surveyed several of the Informix =3D=20 >> architects=3D20=20 >>> and asked them several questions about the evolution of the = Informix=3D20=3D=20 >>=20 >>> technology since the acquisition. I've heard from them. Now I would = =3D=20 >> like=3D20=20 >>> to get your thoughts.=3D20=20 >>> =3D20=20 >>> What do you believe were the top 5 features or enhancements to = the=3D20=20 >>> Informix portfolio since 2001?=3D20=20 >>> =3D20=20 >>> What do you believe were the top 3 under appreciated features=20 >> released=3D20=3D=20 >>=20 >>> over the last 10 years?=3D20=20 >>> =3D20=20 >>> I appreciate your feedback. I will include some of this feedback in=20= >> my=3D20=3D=20 >>=20 >>> keynote next week at the Informix conference in Lenexa and then = again =3D=20 >> in=3D20=20 >>> the Chat with the Lab on the 18th of May.=3D20=20 >>> =3D20=20 >>> Regards,=3D20=20 >>> =3D20=20 >>> Jerry Keesee=3D20=20 >>> World Wide Director, Informix Software=3D20=20 >>> =3D20=20 >>> =3D20=20 >>> =3D=20 >> = **************************************************************************= =3D=20 >> *****=3D20=20 >>> Forum Note: Use "Reply" to post a response in the discussion = forum.=3D20=3D=20 >>=20 >>> =3D20=20 >>=20 >>=20 >>=20 >>=20 > = **************************************************************************= *****=20 >> Forum Note: Use "Reply" to post a response in the discussion forum.=20= >>=20 >>=20 >=20 > --=20 > Fernando Nunes=20 > Portugal=20 >=20 > http://informix-technology.blogspot.com=20 > My email works... but I don't check it frequently...=20 >=20 > --001636c5c22d5e5b7604a31140aa=20 >=20 >=20 > = **************************************************************************= *****=20 > Forum Note: Use "Reply" to post a response in the discussion forum.=20= >=20

↪ replying to Jack Parker

Fernando Nunes — Thu, 12 May 2011 15:08:31 — source: IIUG Forums & Mailing Lists

That cleared a lot of stuff... Some of the things are easy to explain, others not, and could eventually be considered bugs... On Thu, May 12, 2011 at 12:41 PM, Jack Parker <jack.parker4@verizon.net>wrote: > All of the things that you used to be able to do you can still do, it's = > the "new" stuff. > > Try getting HPL to work under windows. There are all sorts of = > restrictions. No pipes. No ipload interface, everything has to go = > ipload needs X-Windows. Windows has no X-Windows... It would be hard to emulate it... > through onpladm, you can't actually invoke the job through onpladm (even = > though it is advertised as such - you have to invoke onpload), I forget = > now, but I seem to remember issues with device arrays. > > > Work on a mac, no Xtree. > > Xtree is deprecated for a long time I believe. As such it wouldn't make sense to put it on mac... > I've just spend a couple of days trying to get ifxdeploy to work on Win = > 2008. You have to turn your body sideways and approach it in just the = > right direction (if you use ifxdeployassist from the command line you = > have to be selective in what options you choose (I've given up there and = > moved to the GUI until I can get ifxdeploy to work), has to have java = > installed and in the PATH, extracts to the local directory instead of = > $INFORMIXDIR) - then complains that %INFORMIXDIR/etc/%ONCONFIG doesn't = > exist - and currently it fails when trying to allocate dbspaces. I'm = > currently working with folks at IBM on this one. > > I havent' work with ifxdeploy. But this may lie under the bug umbrella? > Uninstall on Windows is iffy. You have to check that %INFORMIXDIR is = > actually gone, %PATH doesn't get cleaned up, services may not have been = > cleaned up. I have a list of things I check, and 80% of the time I have = > to manually fix them. > This too... > No fast alter if there's an opaque data type. > > You can't have fast alters that the engine doesn't know how to convert "on the fly". I can understand that this may cause issues, but it's a downside of extensibility... > Building a data management by fragmentation scheme (age off old data by = > dropping fragments). You can't have RI turned on, can't have primary = > keys, I can't recall all of the things I had to work around on that one, = > but it took a couple of months. > Would need more details. And there were some bugs... One limitation is that you can't have a unique index using the same fragment schema if the fragmentation column is not contained in the index. This affects other databases besides Informix and has technical reasons... > There is all this "new" stuff, but it no longer seems to be fully = > thought through. > I suppose you have some valid points, but you also point out things that are simply not possible. I could mention others... raw tables in MACH 11 environments etc. Some things are a natural consequence of using some more advance features. > > I love that building an index updates statistics > > I think we all do :) -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently... --0016e68e80269aff6a04a3158fb5

↪ replying to Fernando Nunes

Madison Pruet — Thu, 12 May 2011 15:34:32 — source: IIUG Forums & Mailing Lists

I've heard the complaint about raw tables before so will address that. MACH11 requires logging to communicate to the secondary nodes what changes have been made on the primary. Yea we could implement a form of 'raw' table support --- but only by internally ignoring the fact that 'raw' tables aren't supposed to be logged and log them anyway. From: "Fernando Nunes" <domusonline@gmail.com> To: ids@iiug.org Date: 05/12/2011 10:11 AM Subject: Re: 10 Years of Innovation at IBM [23675] Sent by: ids-bounces@iiug.org That cleared a lot of stuff... Some of the things are easy to explain, others not, and could eventually be considered bugs... On Thu, May 12, 2011 at 12:41 PM, Jack Parker <jack.parker4@verizon.net>wrote: > All of the things that you used to be able to do you can still do, it's = > the "new" stuff. > > Try getting HPL to work under windows. There are all sorts of = > restrictions. No pipes. No ipload interface, everything has to go = > ipload needs X-Windows. Windows has no X-Windows... It would be hard to emulate it... > through onpladm, you can't actually invoke the job through onpladm (even = > though it is advertised as such - you have to invoke onpload), I forget = > now, but I seem to remember issues with device arrays. > > > Work on a mac, no Xtree. > > Xtree is deprecated for a long time I believe. As such it wouldn't make sense to put it on mac... > I've just spend a couple of days trying to get ifxdeploy to work on Win = > 2008. You have to turn your body sideways and approach it in just the = > right direction (if you use ifxdeployassist from the command line you = > have to be selective in what options you choose (I've given up there and = > moved to the GUI until I can get ifxdeploy to work), has to have java = > installed and in the PATH, extracts to the local directory instead of = > $INFORMIXDIR) - then complains that %INFORMIXDIR/etc/%ONCONFIG doesn't = > exist - and currently it fails when trying to allocate dbspaces. I'm = > currently working with folks at IBM on this one. > > I havent' work with ifxdeploy. But this may lie under the bug umbrella? > Uninstall on Windows is iffy. You have to check that %INFORMIXDIR is = > actually gone, %PATH doesn't get cleaned up, services may not have been = > cleaned up. I have a list of things I check, and 80% of the time I have = > to manually fix them. > This too... > No fast alter if there's an opaque data type. > > You can't have fast alters that the engine doesn't know how to convert "on the fly". I can understand that this may cause issues, but it's a downside of extensibility... > Building a data management by fragmentation scheme (age off old data by = > dropping fragments). You can't have RI turned on, can't have primary = > keys, I can't recall all of the things I had to work around on that one, = > but it took a couple of months. > Would need more details. And there were some bugs... One limitation is that you can't have a unique index using the same fragment schema if the fragmentation column is not contained in the index. This affects other databases besides Informix and has technical reasons... > There is all this "new" stuff, but it no longer seems to be fully = > thought through. > I suppose you have some valid points, but you also point out things that are simply not possible. I could mention others... raw tables in MACH 11 environments etc. Some things are a natural consequence of using some more advance features. > > I love that building an index updates statistics > > I think we all do :) -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently... --0016e68e80269aff6a04a3158fb5 ******************************************************************************* Forum Note: Use "Reply" to post a response in the discussion forum.

↪ replying to Jerry Keesee

VICTOR VAT — Thu, 12 May 2011 15:44:37 — source: IIUG Forums & Mailing Lists

Hello I think the following features are the most better: MACH11 Connection Manager Flexible Grid SQL Administration API CREATE .. IF NOT EXISTS / DROP .. IF EXISTS sql-statement CREATE USER sql-statement With best regard Victor

↪ replying to Madison Pruet

Fernando Nunes — Thu, 12 May 2011 15:50:42 — source: IIUG Forums & Mailing Lists

I understand why they can't be supported. I tried to give this as an example where some features naturally prevent others from working. On Thu, May 12, 2011 at 4:34 PM, Madison Pruet <mpruet@us.ibm.com> wrote: > I've heard the complaint about raw tables before so will address that. > > MACH11 requires logging to communicate to the secondary nodes what changes > have been made on the primary. Yea we could implement a form of 'raw' > table support --- but only by internally ignoring the fact that 'raw' > tables aren't supposed to be logged and log them anyway. > > From: "Fernando Nunes" <domusonline@gmail.com> > > To: ids@iiug.org > > Date: 05/12/2011 10:11 AM > > Subject: Re: 10 Years of Innovation at IBM [23675] > > Sent by: ids-bounces@iiug.org > > That cleared a lot of stuff... > Some of the things are easy to explain, others not, and could eventually be > > considered bugs... > > On Thu, May 12, 2011 at 12:41 PM, Jack Parker > <jack.parker4@verizon.net>wrote: > > > All of the things that you used to be able to do you can still do, it's = > > > the "new" stuff. > > > > Try getting HPL to work under windows. There are all sorts of = > > restrictions. No pipes. No ipload interface, everything has to go = > > > > ipload needs X-Windows. Windows has no X-Windows... It would be hard to > emulate it... > > > through onpladm, you can't actually invoke the job through onpladm (even > = > > though it is advertised as such - you have to invoke onpload), I forget = > > > now, but I seem to remember issues with device arrays. > > > > > > > Work on a mac, no Xtree. > > > > > Xtree is deprecated for a long time I believe. As such it wouldn't make > sense to put it on mac... > > > I've just spend a couple of days trying to get ifxdeploy to work on Win = > > > 2008. You have to turn your body sideways and approach it in just the = > > right direction (if you use ifxdeployassist from the command line you = > > have to be selective in what options you choose (I've given up there and > = > > moved to the GUI until I can get ifxdeploy to work), has to have java = > > installed and in the PATH, extracts to the local directory instead of = > > $INFORMIXDIR) - then complains that %INFORMIXDIR/etc/%ONCONFIG doesn't = > > exist - and currently it fails when trying to allocate dbspaces. I'm = > > currently working with folks at IBM on this one. > > > > > I havent' work with ifxdeploy. But this may lie under the bug umbrella? > > > Uninstall on Windows is iffy. You have to check that %INFORMIXDIR is = > > actually gone, %PATH doesn't get cleaned up, services may not have been = > > > cleaned up. I have a list of things I check, and 80% of the time I have = > > > to manually fix them. > > > > This too... > > > No fast alter if there's an opaque data type. > > > > You can't have fast alters that the engine doesn't know how to convert > "on > the fly". > I can understand that this may cause issues, but it's a downside of > extensibility... > > > Building a data management by fragmentation scheme (age off old data by = > > > dropping fragments). You can't have RI turned on, can't have primary = > > keys, I can't recall all of the things I had to work around on that one, > = > > but it took a couple of months. > > > > Would need more details. And there were some bugs... One limitation is that > > you can't have a unique index using the same fragment schema if the > fragmentation column is not contained in the index. This affects other > databases besides Informix and has technical reasons... > > > There is all this "new" stuff, but it no longer seems to be fully = > > thought through. > > > > I suppose you have some valid points, but you also point out things that > are > simply not possible. I could mention others... raw tables in MACH 11 > environments etc. > Some things are a natural consequence of using some more advance features. > > > > > I love that building an index updates statistics > > > > > I think we all do :) > -- > Fernando Nunes > Portugal > > http://informix-technology.blogspot.com > My email works... but I don't check it frequently... > > --0016e68e80269aff6a04a3158fb5 > > > > ******************************************************************************* > > Forum Note: Use "Reply" to post a response in the discussion forum. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > -- Fernando Nunes Portugal http://informix-technology.blogspot.com My email works... but I don't check it frequently... --001636ef0da8747d6604a31626e9

↪ replying to Fernando Nunes

Jack Parker — Thu, 12 May 2011 16:08:32 — source: IIUG Forums & Mailing Lists

Windows is an abomination and it's a wonder that anything works on it. = I really have no right to whine. SW is also more complicated and the stack is deeper than it was 20 years = ago, any time we cobble stuff together there are going to be things that = don't fit cleanly. It would be nice to have some sort of HPL GUI on windows - even if it's = not X related. However, it's not worth the trouble, onpladm pretty much = does the job. Xtree was a wonderful tool and it's a shame that it's been abandoned. Perhaps I'm working at a lower level than I was 20 years ago, it just = seems that I run into more trouble getting stuff to work these days. My = first day on unix, I figured out lpadmin in under an hour. Decades = later I'm struggling with ifxdeploy. cheers j. On May 12, 2011, at 11:50 AM, Fernando Nunes wrote: > I understand why they can't be supported. I tried to give this as an = example=20 > where some features naturally prevent others from working.=20 >=20 > On Thu, May 12, 2011 at 4:34 PM, Madison Pruet <mpruet@us.ibm.com> = wrote:=20 >=20 >> I've heard the complaint about raw tables before so will address = that.=20 >>=20 >> MACH11 requires logging to communicate to the secondary nodes what = changes=20 >> have been made on the primary. Yea we could implement a form of 'raw'=20= >> table support --- but only by internally ignoring the fact that 'raw'=20= >> tables aren't supposed to be logged and log them anyway.=20 >>=20 >> From: "Fernando Nunes" <domusonline@gmail.com>=20 >>=20 >> To: ids@iiug.org=20 >>=20 >> Date: 05/12/2011 10:11 AM=20 >>=20 >> Subject: Re: 10 Years of Innovation at IBM [23675]=20 >>=20 >> Sent by: ids-bounces@iiug.org=20 >>=20 >> That cleared a lot of stuff...=20 >> Some of the things are easy to explain, others not, and could = eventually be=20 >>=20 >> considered bugs...=20 >>=20 >> On Thu, May 12, 2011 at 12:41 PM, Jack Parker=20 >> <jack.parker4@verizon.net>wrote:=20 >>=20 >>> All of the things that you used to be able to do you can still do, = it's =3D=20 >>=20 >>> the "new" stuff.=20 >>>=20 >>> Try getting HPL to work under windows. There are all sorts of =3D=20 >>> restrictions. No pipes. No ipload interface, everything has to go =3D=20= >>>=20 >>=20 >> ipload needs X-Windows. Windows has no X-Windows... It would be hard = to=20 >> emulate it...=20 >>=20 >>> through onpladm, you can't actually invoke the job through onpladm = (even=20 >> =3D=20 >>> though it is advertised as such - you have to invoke onpload), I = forget =3D=20 >>=20 >>> now, but I seem to remember issues with device arrays.=20 >>>=20 >>>=20 >>=20 >>> Work on a mac, no Xtree.=20 >>>=20 >>>=20 >> Xtree is deprecated for a long time I believe. As such it wouldn't = make=20 >> sense to put it on mac...=20 >>=20 >>> I've just spend a couple of days trying to get ifxdeploy to work on = Win =3D=20 >>=20 >>> 2008. You have to turn your body sideways and approach it in just = the =3D=20 >>> right direction (if you use ifxdeployassist from the command line = you =3D=20 >>> have to be selective in what options you choose (I've given up there = and=20 >> =3D=20 >>> moved to the GUI until I can get ifxdeploy to work), has to have = java =3D=20 >>> installed and in the PATH, extracts to the local directory instead = of =3D=20 >>> $INFORMIXDIR) - then complains that %INFORMIXDIR/etc/%ONCONFIG = doesn't =3D=20 >>> exist - and currently it fails when trying to allocate dbspaces. I'm = =3D=20 >>> currently working with folks at IBM on this one.=20 >>>=20 >>>=20 >> I havent' work with ifxdeploy. But this may lie under the bug = umbrella?=20 >>=20 >>> Uninstall on Windows is iffy. You have to check that %INFORMIXDIR is = =3D=20 >>> actually gone, %PATH doesn't get cleaned up, services may not have = been =3D=20 >>=20 >>> cleaned up. I have a list of things I check, and 80% of the time I = have =3D=20 >>=20 >>> to manually fix them.=20 >>>=20 >>=20 >> This too...=20 >>=20 >>> No fast alter if there's an opaque data type.=20 >>>=20 >>> You can't have fast alters that the engine doesn't know how to = convert=20 >> "on=20 >> the fly".=20 >> I can understand that this may cause issues, but it's a downside of=20= >> extensibility...=20 >>=20 >>> Building a data management by fragmentation scheme (age off old data = by =3D=20 >>=20 >>> dropping fragments). You can't have RI turned on, can't have primary = =3D=20 >>> keys, I can't recall all of the things I had to work around on that = one,=20 >> =3D=20 >>> but it took a couple of months.=20 >>>=20 >>=20 >> Would need more details. And there were some bugs... One limitation = is that=20 >>=20 >> you can't have a unique index using the same fragment schema if the=20= >> fragmentation column is not contained in the index. This affects = other=20 >> databases besides Informix and has technical reasons...=20 >>=20 >>> There is all this "new" stuff, but it no longer seems to be fully =3D=20= >>> thought through.=20 >>>=20 >>=20 >> I suppose you have some valid points, but you also point out things = that=20 >> are=20 >> simply not possible. I could mention others... raw tables in MACH 11=20= >> environments etc.=20 >> Some things are a natural consequence of using some more advance = features.=20 >>=20 >>>=20 >>> I love that building an index updates statistics=20 >>>=20 >>>=20 >> I think we all do :)=20 >> --=20 >> Fernando Nunes=20 >> Portugal=20 >>=20 >> http://informix-technology.blogspot.com=20 >> My email works... but I don't check it frequently...=20 >>=20 >> --0016e68e80269aff6a04a3158fb5=20 >>=20 >>=20 >>=20 >>=20 > = **************************************************************************= *****=20 >>=20 >> Forum Note: Use "Reply" to post a response in the discussion forum.=20= >>=20 >>=20 >>=20 >>=20 > = **************************************************************************= *****=20 >> Forum Note: Use "Reply" to post a response in the discussion forum.=20= >>=20 >>=20 >=20 > --=20 > Fernando Nunes=20 > Portugal=20 >=20 > http://informix-technology.blogspot.com=20 > My email works... but I don't check it frequently...=20 >=20 > --001636ef0da8747d6604a31626e9=20 >=20 >=20 > = **************************************************************************= *****=20 > Forum Note: Use "Reply" to post a response in the discussion forum.=20= >=20

↪ replying to VICTOR VAT

Art Kagel — Thu, 12 May 2011 16:12:12 — source: IIUG Forums & Mailing Lists

Related to CREATE IF NOT EXISTS/DROP IF EXISTS... We could use a CREATE OR REPLACE option. The IF EXISTS logic will eliminate the error messages when scripting a table replace/create script, but it's still not as nice as CREATE OR REPLACE for any object. Art Art S. Kagel Advanced DataTools (www.advancedatatools.com) Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on my employer, Advanced DataTools, the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. 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 Thu, May 12, 2011 at 11:44 AM, VICTOR VAT <victor16@inbox.ru> wrote: > Hello > > I think the following features are the most better: > > MACH11 > Connection Manager > Flexible Grid > SQL Administration API > CREATE .. IF NOT EXISTS / DROP .. IF EXISTS sql-statement > CREATE USER sql-statement > > With best regard > Victor > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --20cf3071d02e8f05e004a31673a9

↪ replying to Art Kagel

VICTOR VAT — Thu, 12 May 2011 16:37:05 — source: IIUG Forums & Mailing Lists

We can achieve the same result with CREATE OR REPLACE for couple statements: DROP .. IF EXISTS CREATE ... I mentioned these additions in the sense that without one the previous versions returned the error, which was a big headache for programmers. But as for us programmers without a difference how to call it :).

↪ replying to Jerry Keesee

FRANK — Fri, 13 May 2011 21:27:23 — source: IIUG Forums & Mailing Lists

Hello, --MACH11 ( ER, RSS...) -- Data Compression -- Single User mode -- Query optimization -- Buffer management, multiple page size, non-blocking check point Thanks, Frank On Tue, May 10, 2011 at 4:05 PM, Jerry Keesee <jkeesee@us.ibm.com> wrote: > July 1st will be the 10th anniversary of the IBM acquisition of the > Informix assets. I recently surveyed several of the Informix architects > and asked them several questions about the evolution of the Informix > technology since the acquisition. I've heard from them. Now I would like > to get your thoughts. > > What do you believe were the top 5 features or enhancements to the > Informix portfolio since 2001? > > What do you believe were the top 3 under appreciated features released > over the last 10 years? > > I appreciate your feedback. I will include some of this feedback in my > keynote next week at the Informix conference in Lenexa and then again in > the Chat with the Lab on the 18th of May. > > Regards, > > Jerry Keesee > World Wide Director, Informix Software > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --bcaec5430b62721ea704a32ef8dd