[firebird-support] Re: Off-Topic: Firebird future

2019-10-09 Thread pablo sanchez pab...@adinet.com.uy [firebird-support]
I use Firebird since 0.x version (open interbase?) .

It's the most versatile db I've found so far : embedded and full server,
linux(99%) and windows, odbc or jaybird.

Installs in 5 minutes with db configuration.

I never had a performance issue that was not my fault and biggest db's
are 80GB+ with ~300 concurrent sessions .

The "tooling" is becoming a problem for me as I use less and less
ms-windows. There are no linux tools I can bare to use daily as 
Flamerobin is, and I must start migrating to v3 ASAP. I'm starting to 
evaluate windows tools within WINE and java based tools .

Regards

Pablo Sánchez

> Hello everyone,
> 
> 
> 
> I've been using FB since version 1 and I have only good comments about 
> the database quality and performance. However, It's true that the 
> development has evolved too slow and we had to wait almost 10 years 
> since the first release to get a real SMP support (I know that's tricky 
> since for Linux and classic was not a problem) and we were far behind 
> the competitors MySQL and PgSQL.
> 
> In the meantime the great team was improving the database, MySQL and 
> PgSQL have won a lot of user base because they evolved faster adding new 
> features like XML support, JSON, NoSQL engines, etc. into their database 
> core. Some of those features look far away from the Firebird next versions.
> 
> Also, today I was looking for a Flamerobin updated version but the 
> project seens abandoned. I mean, flamerobin is a really cool tool but 
> nobody is helping its developers to keep it updated. Some of you can 
> blame that everyone should be able to create code for the project, but 
> that's not true, a project needs engagement to develop, knowledge of the 
> language, etc. So it's not e asy to add developers (and more difficult 
> if they don't see cool features out of the box). Basically, if you want 
> a management tool updated you need to go to commercial.
> 
> So, my point is, the current Firebird develepment model is not 
> delivering features as the market requires, there is not a gui 
> management tool once you install the software, all the cool features are 
> commercial. If you install MySQL, after the installation you will get a 
> DBMS with GUI for management, encryption,  replication out of the box 
> since many years ago, all drivers for developers if desired, extensive 
> support from Cloud providers, and a lot of features that make it a very 
> functional tool out of the box. Also PgSQL offers many of these features 
> out of the box fo r free.
> 
> I know the Firebird model offers to many Companies to sell those missing 
> tools, but to be honest, if your competitors offer the same for free you 
> will go for them. In my last project I used MySQL because it has wide 
> support in the cloud providers, offering Database as a Service, 
> replication, encryption, an integrated GUI out of the box and other 
> features. In the next project I will be using PgSQL for similar reasons. 
> I'm asking myself how much time I'll still be using Firebird if the 
> project evolves slow, there is not GUI officially maintened and the cool 
> features are for the commercial companies.
> 
> I don't want to start a flame, thos e are only my personal thoughs 
> regarding the Firebird future. If nothing changes, we could see some 
> users and developers moving to other DBMS for sure... and we have not 
> talked about NoSQL competitors yet...
> 
> You are the genius, I hope you will figure out how to keep this great 
> project alive.
> 
> 
> 
> 
> Best regards,
> 
> 
> 




Re: [firebird-support] Plugin(?) for external tabels in CSV or Excel?

2019-10-09 Thread Dimitry Sibiryakov s...@ibphoenix.com [firebird-support]
09.10.2019 17:24, Kjell Rilbe kjell.ri...@marknadsinformation.se 
[firebird-support] wrote:
> Anyway, would it be possible to write some kind of plugin to FB 3 to
> allow it to select from or insert into either CSV or Excel files or
> both? I would code the actual plugin myself, of course. I'm just asking
> if it's worth investigating (not waste time if it's a known dead end).

   The main problem with this is fixed metadata for every database object. If 
it is fine 
for your purpose, you can start from coding UDR returning result set from a 
file of any 
(including CSV) format.


-- 
   WBR, SD.






++

Visit http://www.firebirdsql.org and click the Documentation item
on the main (top) menu.  Try FAQ and other links from the left-side menu there.

Also search the knowledgebases at http://www.ibphoenix.com/resources/documents/ 

++


Yahoo Groups Links

<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/firebird-support/

<*> Your email settings:
Individual Email | Traditional

<*> To change settings online go to:
http://groups.yahoo.com/group/firebird-support/join
(Yahoo! ID required)

<*> To change settings via email:
firebird-support-dig...@yahoogroups.com 
firebird-support-fullfeatu...@yahoogroups.com

<*> To unsubscribe from this group, send an email to:
firebird-support-unsubscr...@yahoogroups.com

<*> Your use of Yahoo Groups is subject to:
https://info.yahoo.com/legal/us/yahoo/utos/terms/



[firebird-support] Plugin(?) for external tabels in CSV or Excel?

2019-10-09 Thread Kjell Rilbe kjell.ri...@marknadsinformation.se [firebird-support]
Hi,



We work in .NET and C# and we often import and export Excel or CSV data. 

It would be really good to be able to select from and insert (append) 

directly into an external CSV file or Excel file.



I am aware of the currently supported external table feature, but the 

format is rather difficult to use, especially for non-char data nd for 

char data in encoding with variable length characters (e.g. UTF8). I 

have written a tool to create such files (with UTF8 strings), but it's a 

free standing app and I need to prepare the file first, then use the 

tools generated DDL to create the external table within Firebird, before 

I can actually access the data from SQL.



I know new FB verions has support for plugins and as far as I understand 

even a separate engine could be plugged in, right? (I may be wrong here, 

haven't read up on it properly...)



Anyway, would it be possible to write some kind of plugin to FB 3 to 

allow it to select from or insert into either CSV or Excel files or 

both? I would code the actual plugin myself, of course. I'm just asking 

if it's worth investigating (not waste time if it's a known dead end).



If it is indeed possible, I'd appreaciate some pointer how to get 

started: relevant docs or some sample code, or whatever.



If I'm successful I'd be happy to open source it, but if I am to write 

it, it will be C#/.NET (too much effort for me otherwise).



Thanks,

Kjell





[Non-text portions of this message have been removed]



Re: [firebird-support] Scaling Firebird - Azure

2019-10-09 Thread 'Alexey Kovyazin (ak)' a...@ib-aid.com [firebird-support]
Hello,

Azure simply does not deliver what it promises - simple test
www.ib-aid.com/dbtest gives results for Azure premium ssd in the lowest
25%, simple CrystalDiskMarks shows awful results too.
Your server is yours, but Azure promises Enterprise grade performance,
without worries abour RAID, ssd, etc - but what we get is far from it, as
well as from money paid for it. On other side, Google Cloud shows
reasonable results in these tests, so why pay for the worst option.
Amazon EC2 with ssd also looks better than Azure in terms of performance,
but seems to be much more expensive than Digital Ocean, Hetzner and even
Google Cloud.



Regards,
Alexey Kovyazin
IBSurgeon

ср, 9 окт. 2019 г., 10:49 'Louis van Alphen' lo...@nucleo.co.za
[firebird-support] :

>
>
> I think it is unfair to say Azure sucks. I have had huge performance
> issues with FB on a normal server with server grade SSDs. My desktop PC
> with spindle HDD outperforms it by an order of magnitude. To date it has
> not been resolved but I suspect it has to do with the RAID controller. The
> same server is very fast with MSSQL.
>
> From: firebird-support@yahoogroups.com [mailto:
> firebird-supp...@yahoogroups..com]
> Sent: Wednesday, 09 October, 2019 08:17
> To: Nagy Szilveszter nagy_szilvesz...@yahoo.com [firebird-support] <
> firebird-support@yahoogroups.com>
> Subject: Re: [firebird-support] Scaling Firebird - Azure
>
> Hello,
>
> In short words, Azure sucks, its so called Premium SSD is worse than the
> cheapest consumer grade ssd.
>
> Try Google Cloud or Amazon, or Hetzner.
>
> Regards,
>
> Alexey Kovyazin
>
> IBSurgeon
>
> пн, 7 окт. 2019 г., 0:20 Rune Horneland rune.hornel...@kravia.net  rune.hornel...@kravia.net> [firebird-support] <
> firebird-support@yahoogroups.com 
> >:
>
> Hi,
>
> We are running Firebird 2 in an Azure VM. It can only take so much in
> terms of concurrent connections.
>
> What top-level advice would you give to scale this?
>
> We are connecting to it from a .NET core Middleware using Azure VMs.
>
> The architecture of the middleware is quite monolithic. We are considering
> rewriting it with Microservices and Azure functions or similar architecure,
> but are unsure how we could scale the Firebird DB or connections itself.
>
> Multiple casehandlers in our company use it via a Delphi-based Windows
> application at the other end, with a vendor maintaining the Firebird DB and
> Windows app development, so we are locked into using Firebird.
>
> RUNE HORNELAND
> CTO
>
> [Non-text portions of this message have been removed]
>
> 
>


RE: [firebird-support] Scaling Firebird - Azure

2019-10-09 Thread ma_go...@yahoo.com [firebird-support]
Software raid and hardware raid without BBU has a huge performace hit :

 https://ib-aid.com/en/articles/45-ways-to-speed-up-firebird-database/ 
https://ib-aid.com/en/articles/45-ways-to-speed-up-firebird-database/

 

 EXT3, EXT4 : file system barrier property has huge performance hit
 https://www.firebirdnews.org/understanding-barrier-on-linux/ 
https://www.firebirdnews.org/understanding-barrier-on-linux/

 https://www.firebirdnews.org/forced-writes-performance-impact-on-ubuntu/ 
https://www.firebirdnews.org/forced-writes-performance-impact-on-ubuntu/

 

 Hyperthreading could harm performance : 
https://blogs.msdn.microsoft.com/slavao/2005/11/12/be-aware-to-hyper-or-not-to-hyper/
 
https://blogs.msdn.microsoft.com/slavao/2005/11/12/be-aware-to-hyper-or-not-to-hyper/


RE: [firebird-support] Scaling Firebird - Azure

2019-10-09 Thread 'Louis van Alphen' lo...@nucleo.co.za [firebird-support]
I think it is unfair to say Azure sucks. I have had huge performance issues 
with FB on a normal server with server grade SSDs. My desktop PC with spindle 
HDD outperforms it by an order of magnitude. To date it has not been resolved 
but I suspect it has to do with the RAID controller. The same server is very 
fast with MSSQL. 

 

From: firebird-support@yahoogroups.com 
[mailto:firebird-supp...@yahoogroups..com] 
Sent: Wednesday, 09 October, 2019 08:17
To: Nagy Szilveszter nagy_szilvesz...@yahoo.com [firebird-support] 

Subject: Re: [firebird-support] Scaling Firebird - Azure

 

  

Hello,

 

In short words, Azure sucks, its so called Premium SSD is worse than the 
cheapest consumer grade ssd.

 

Try Google Cloud or Amazon, or Hetzner.

 

Regards, 

Alexey Kovyazin 

IBSurgeon 

 

 

 

пн, 7 окт. 2019 г., 0:20 Rune Horneland rune.hornel...@kravia.net 
  [firebird-support] 
mailto:firebird-support@yahoogroups.com> >:

  

Hi,

We are running Firebird 2 in an Azure VM. It can only take so much in terms of 
concurrent connections.

What top-level advice would you give to scale this?


We are connecting to it from a .NET core Middleware using Azure VMs.

The architecture of the middleware is quite monolithic. We are considering 
rewriting it with Microservices and Azure functions or similar architecure, but 
are unsure how we could scale the Firebird DB or connections itself.

Multiple casehandlers in our company use it via a Delphi-based Windows 
application at the other end, with a vendor maintaining the Firebird DB and 
Windows app development, so we are locked into using Firebird. 

 

RUNE HORNELAND
CTO





[Non-text portions of this message have been removed]



Re: [firebird-support] Off-Topic: Firebird future

2019-10-09 Thread liviuslivius liviusliv...@poczta.onet.pl [firebird-support]
Hi
 
lack of GUI tool is a realy big problem, especially for newcommers.
As you have pointed flamerobin looks like abandoned project. It have some 
problems but was quite good tool.
 
But to compare Firebird to other servers like Postgress few points must be 
taken.
1. Firebird have simple instalation and its size is really small. It is simple 
to administration by novice users.
2. Firebird database file is single flat file. It is simple to backup. In 
Postgress it is some folder with number in the name and many files. Novice user 
do not know how to backup it.
3. Firebird have genius embeded version. I use it on near every webpage as i do 
not depend on preinstalled version on the provider side. I am also not limited 
in size of my database :)
   It is working on medium trust servers :)
4. I have made comparission in speed for my biggest project, with not so big 
database file. It is only 50GB. I have catch near the all sql from the system. 
And the resutls are:
   66% of sql are run faster on Postgress 34% are run faster on Firebird. One 
critical for us sql was run faster on Firebird side and it is winner for us 
here.
   I know this is specific comparisoin but i have analysed where the problems 
are.
   I describe it in next points (5..9)
5. Firebird do not have index only scans and include clause in index. This is 
the biggest lack. Because of this i have migreted ~800GB database, 
noncommercial project based on GAIA DR2 data from Firebird. Look at 
http://tracker.firebirdsql.org/browse/CORE-5909.
6. Firebird do not have parallel operations.
7. Firebird indexes are not bidirectional. You must specify asc desc. (this is 
not so big problem as you can crate both separately).
8. Firebird compound indexes cannot use parts from composite index if middle 
part is missed. E.g. 1,2,4 but 3 is missed. Postgres catch this.
9. Lack of hash/merge algorithms for outer joins. Also believe that it will be 
in FB4 :)
10. Firebird triggers and procedures are really good. I know that newcomers 
have problems with it i do not know why. Maybe something obvious is missed in 
the documentation.
11. Firebird miss some obvious feature which made me crasy. Import from CSV 
file. I have proposed simple fix for this 
http://tracker.firebirdsql.org/browse/CORE-5818 but it was rejected by the team 
as the more common/extended version is needed.
    I understand this, but more extended version will be implemented maybe in 
next 10 years but my simple proposition i suppose can be implemented in few 
days. I know it must be then maitained when more extended version will be 
avaiable. This is a know every day developer problem i understand.
12. Lack of lateral join, i believe that it will be in FB4 :)
13. Firebird have really good system monitoring. It have good trace and MON$ 
tables. I like it very much.
14. Firebird have non detailed info in query plan. Compared to other servers.
15. Firebird lack of running query plan as it depend strongly on plan gethered 
at prepare time. It cannot addapt to value provided in parameter by histogram 
use.
16. Firebird have really good CTE also recursive where you can provide almost 
every feature avaiable in normal sql.
17. Firebird have quite good support. But i suppose, team do not have dedicated 
person for this point. Compared to Interbase it is still much much better :) IB 
paid support is something what should be banned. We have migrated all projects 
from Interbase to Firebird because of this.
18. Lack of grouping sets.
19. Finally Firebird 3 is really stable for us. No problems occurred (fingers 
crossed). This is something what also cannot be missed.
 
Finally i can only say really good words about Firebird team. With such limited 
resources they do really big progress. Compare how many persons are in Firebird 
team and how many in Postgress.
The problem is that team is overloaded and new peresons also when provide pull 
requests cannot be carried by the hand from the team as team have not time and 
circle is closed.
I see that some users provide pull requests and as they looks finished stay not 
merged. 
This should be improved as such person can provide more pull requests and maybe 
in the future will be in the team as a member. Maybe not, but every brick 
counts.
E.g. https://github.com/FirebirdSQL/firebird/pull/204
 
I greet everyone warmly and i hope that the Firebird team will grow in the 
future
Karol Bieniaszewski
 

Re: [firebird-support] Scaling Firebird - Azure

2019-10-09 Thread 'Alexey Kovyazin (ak)' a...@ib-aid.com [firebird-support]
Hello,

In short words, Azure sucks, its so called Premium SSD is worse than the
cheapest consumer grade ssd.

Try Google Cloud or Amazon, or Hetzner.

Regards,
Alexey Kovyazin
IBSurgeon




пн, 7 окт. 2019 г., 0:20 Rune Horneland rune.hornel...@kravia.net
[firebird-support] :

>
>
> Hi,
> We are running Firebird 2 in an Azure VM. It can only take so much in
> terms of concurrent connections.
> What top-level advice would you give to scale this?
>
> We are connecting to it from a .NET core Middleware using Azure VMs.
> The architecture of the middleware is quite monolithic. We are considering
> rewriting it with Microservices and Azure functions or similar architecure,
> but are unsure how we could scale the Firebird DB or connections itself.
> Multiple casehandlers in our company use it via a Delphi-based Windows
> application at the other end, with a vendor maintaining the Firebird DB and
> Windows app development, so we are locked into using Firebird.
>
>
> *RUNE HORNELAND*CTO
>
> 
>