Diese Seite mit anderen teilen ...

Informationen zum Thema:
Forum:
WinDev Forum
Beiträge im Thema:
10
Erster Beitrag:
vor 6 Monaten, 1 Woche
Letzter Beitrag:
vor 6 Monaten
Beteiligte Autoren:
PETER ZHOU, Fabrice Harari, Arie, ccc2

Using hyperfile SQL for millions of records

Startbeitrag von PETER ZHOU am 14.02.2017 04:30

Hi,

Anybody having experience with hyperfile SQL with memo tables having millions of records?

Any issue ? Slow ? Corruption ?

Regards,

PETER ZHOU

Antworten:

Hi

>>Anybody having experience with hyperfile SQL with memo tables having millions of records?

Yes...

>>Any issue ?

No

>>Slow ?

No (my analysis are generally correctly structured, thank you)

>>Corruption ?

No

Best regards

von Fabrice Harari - am 14.02.2017 11:59
Quote
Fabrice Harari
Hi

>>Anybody having experience with hyperfile SQL with memo tables having millions of records?

Yes...

>>Any issue ?

No

>>Slow ?

No (my analysis are generally correctly structured, thank you)

>>Corruption ?

No

Best regards


Hi Fabrice,

Do you know hyperfile is based on what kind of file format standards ?

von PETER ZHOU - am 14.02.2017 12:22
Hi again

1. I don't know what you mean by "file format standards"
2. As soon as I hear the word "standard" in association with any type of computer stuff, I always hesitate between laughing and crying.

Best regards

von Fabrice Harari - am 14.02.2017 13:00
Quote
Fabrice Harari
Hi again

1. I don't know what you mean by "file format standards"
2. As soon as I hear the word "standard" in association with any type of computer stuff, I always hesitate between laughing and crying.

Best regards


Hi Fabrice,

Sorry for not being clear on my question :)

My client asking as they want to know the architecture of hyperfile.
for example they can search for Btrieve architecture but not hyperfile.

Regards,

PETER ZHOU

von PETER ZHOU - am 14.02.2017 13:14
Hi Peter,

hyperfile uses it's own internal structure for storing data. Every database-table is a file on disk, like with dBase. You can use them in classic-mode, accessing those files directly, again like in dBase. Or in client-server mode (like mssql,myssql and so on) where there is a database service between your application and the physical files.
Millions of records with images - no problem. Regular data is stored in a datafile (ie tbl_customers.fic) . Indexes are stored in a spearate file (tbl_customers.ndx) and memo-fields are stored as a pointer in the datafile. The actual memodata is stored in a 3rd file with the extension mmo. (tbl_customers.mmo). This one get's big with millions of records.

von Arie - am 14.02.2017 13:16
Quote
Arie
Hi Peter,

hyperfile uses it's own internal structure for storing data. Every database-table is a file on disk, like with dBase. You can use them in classic-mode, accessing those files directly, again like in dBase. Or in client-server mode (like mssql,myssql and so on) where there is a database service between your application and the physical files.
Millions of records with images - no problem. Regular data is stored in a datafile (ie tbl_customers.fic) . Indexes are stored in a spearate file (tbl_customers.ndx) and memo-fields are stored as a pointer in the datafile. The actual memodata is stored in a 3rd file with the extension mmo. (tbl_customers.mmo). This one get's big with millions of records.


Hi Arie,

Thank you for the information. I do understand these points.
Let's take dBase, I think dBASE becomes slow when it hits millions of records
How is hyperfile different from dBASE? Since both of them are based on ISAM.


Regards,

PETER ZHOU

von PETER ZHOU - am 14.02.2017 13:27
Peter,

I don't know how the indexes are stuctured, B-tree or otherwise. Only pcsoft knows that.
From personal experience I can tell that performance has never been an issue to me. Which means no guarantee for your situation, I'm afraid.

On the other hand: Google for mssql or Oracle and performance and you will get tons of issues too. It's always mix of application design and databasedesign.

von Arie - am 14.02.2017 14:28
Quote
Arie
Peter,

On the other hand: Google for mssql or Oracle and performance and you will get tons of issues too. It's always mix of application design and databasedesign.


Arie,

The drawback for hyperfile SQL
1. can only work with simple sql
2. not sql compliance , good luck with hyperfile SQLnative sql languages
3. when executing complex sql , hyperfile SQL will try break it into simple sql which causing performance issue.
4. ..... etc ...

mssql or Oracle may has performance issues but they can do a tons more stuff than hyperfile SQL


Peter,

if you looking working on small program (no complex sql) then hyperfile SQL is good choice but if we talking about 100 tables and 30GB size then you should use oracle , mssql or mysql (PL-sql > T-sql > sql)

use hyperfile SQL only if you really really have no choice .

von ccc2 - am 17.02.2017 11:01
Hi

>>if you looking working on small program (no complex sql) then hyperfile SQL is good choice but if we talking about 100 tables and 30GB size then you should use oracle , mssql or mysql (PL-sql > T-sql > sql)

use hyperfile SQL only if you really really have no choice .

von Fabrice Harari - am 17.02.2017 14:54
Zur Information:
MySnip.de hat keinen Einfluss auf die Inhalte der Beiträge. Bitte kontaktieren Sie den Administrator des Forums bei Problemen oder Löschforderungen über die Kontaktseite.
Falls die Kontaktaufnahme mit dem Administrator des Forums fehlschlägt, kontaktieren Sie uns bitte über die in unserem Impressum angegebenen Daten.