| View previous topic :: View next topic |
| Author |
Message |
shotokan99 Guest
|
Posted: Tue Jul 25, 2006 8:08 am Post subject: for flat-file application..which is faster to process data t |
|
|
guys pls i need ur input |
|
| Back to top |
|
 |
DB-1 Guest
|
Posted: Tue Jul 25, 2006 2:52 pm Post subject: Re: for flat-file application..which is faster to process da |
|
|
shotokan99 wrote:
| Quote: |
guys pls i need ur input
|
If you really do, then re-phrase your posting. And put *much* more
effort in to those sentences when you describe your situation.
"guys pls i need ur input", I see no question in your message!!!
NEVER put your question in the Header only, that's considered very bad
news behaviour. The actual question must always be represented in the
body of the message also. |
|
| Back to top |
|
 |
Bruce Roberts Guest
|
Posted: Tue Jul 25, 2006 10:42 pm Post subject: Re: for flat-file application..which is faster to process da |
|
|
"shotokan99" <soft_devjava (AT) yahoo (DOT) com> wrote in message
news:1153796909.965220.221250 (AT) i3g2000cwc (DOT) googlegroups.com...
| Quote: | guys pls i need ur input
|
Who is faster John Smith or Jack Winter? |
|
| Back to top |
|
 |
shotokan99 Guest
|
Posted: Wed Jul 26, 2006 5:18 am Post subject: Re: for flat-file application..which is faster to process da |
|
|
ah ok tnx for the tip. im writing a stand alone application now im
deciding either to use txtfile or dbf files as my database. after all
the data to be stored wont be updated that much like what is done with
rdms tables.
most likely the data will be gathered just for report purposes only. so
which is faster to use between the two it terms of data processing? |
|
| Back to top |
|
 |
shotokan99 Guest
|
Posted: Wed Jul 26, 2006 8:35 am Post subject: Re: for flat-file application..which is faster to process da |
|
|
guys i have this scenario:
from dbf files containing around 300k records im going to replicate
these data using txtfile..some sort like of a backup. then im going to
repeat the same process on daily basis.
which is better to use txtfile or dbf file? which is faster? |
|
| Back to top |
|
 |
Jamie Guest
|
Posted: Wed Jul 26, 2006 8:35 am Post subject: Re: for flat-file application..which is faster to process da |
|
|
shotokan99 wrote:
| Quote: | guys pls i need ur input
it depends on how large of a file your |
working with?
--
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5 |
|
| Back to top |
|
 |
Jamie Guest
|
Posted: Wed Jul 26, 2006 8:35 am Post subject: Re: for flat-file application..which is faster to process da |
|
|
shotokan99 wrote:
| Quote: | ah ok tnx for the tip. im writing a stand alone application now im
deciding either to use txtfile or dbf files as my database. after all
the data to be stored wont be updated that much like what is done with
rdms tables.
most likely the data will be gathered just for report purposes only. so
which is faster to use between the two it terms of data processing?
if its just soming like appending data at the end of the file as you |
add to it until at which point your read to examine it., then
a Text file would be the best and easiest way..
--
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5 |
|
| Back to top |
|
 |
DB-1 Guest
|
Posted: Wed Jul 26, 2006 1:36 pm Post subject: Re: for flat-file application..which is faster to process da |
|
|
shotokan99 wrote:
| Quote: |
from dbf files containing around 300k records im going to replicate
these data using txtfile..some sort like of a backup. then im going to
repeat the same process on daily basis.
|
You seem to have lost the Shift key in your keyboard. And you really are
not giving too much details about your problem. So let's start this
frustrating quessing work...
You have 300.000 records, each consisting about (?) 100 bytes, so the
source DBF file size is something like 30 MB? If you just want to
replicate the whole DBF file, why not just copy it somewhere?
| Quote: | which is better to use txtfile or dbf file? which is faster?
|
Or maybe for some reason you have decided to read through all the 300k
records, make some comparison for each of them, and then write the
selected records out, to some backup file. And for each day you create a
new backup file for the output, without appending the data to the
yesterday's backup file.
If that's the case, then there should not be any faster method to store
the data to a file than use Delphi's TStreams. DBF file format can't be
any faster when doing this kind of raw writing.
Yet your possible problem may lie totally elsewhere. Maybe the bottle
neck is in the reading part, looping through all those 300 records.
And you forgot to mention why it is so important to get this done so
quickly? Maybe you could just copy the original DBF file to some
temporary DBF file. Quickly put the original DBF back to it's daily
work, and then spend the rest of the day manipulating and sucking the
backup data from that temporary file.
DB-1 |
|
| Back to top |
|
 |
Jamie Guest
|
Posted: Thu Jul 27, 2006 6:40 am Post subject: Re: for flat-file application..which is faster to process da |
|
|
shotokan99 wrote:
| Quote: | guys i have this scenario:
from dbf files containing around 300k records im going to replicate
these data using txtfile..some sort like of a backup. then im going to
repeat the same process on daily basis.
which is better to use txtfile or dbf file? which is faster?
text files are fine for backing up data of that size but not |
for random access.
--
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5 |
|
| Back to top |
|
 |
pr Guest
|
Posted: Fri Jul 28, 2006 7:45 pm Post subject: Re: for flat-file application..which is faster to process da |
|
|
"shotokan99" <soft_devjava (AT) yahoo (DOT) com> wrote in message
news:1153884952.801570.7180 (AT) s13g2000cwa (DOT) googlegroups.com...
| Quote: | guys i have this scenario:
from dbf files containing around 300k records im going to replicate
these data using txtfile..some sort like of a backup. then im going to
repeat the same process on daily basis.
which is better to use txtfile or dbf file? which is faster?
|
Is the data read only? If not:
-How are you going to update record 285000 in the text file?
-How are you going to find a record in the text file?
Why not just backup the dbf files?
PR |
|
| Back to top |
|
 |
pr Guest
|
Posted: Sat Jul 29, 2006 1:00 am Post subject: Re: for flat-file application..which is faster to process da |
|
|
"shotokan99" <soft_devjava (AT) yahoo (DOT) com> wrote in message
news:1153873133.625423.272320 (AT) m73g2000cwd (DOT) googlegroups.com...
| Quote: | ah ok tnx for the tip. im writing a stand alone application now im
deciding either to use txtfile or dbf files as my database. after all
the data to be stored wont be updated that much like what is done with
rdms tables.
most likely the data will be gathered just for report purposes only. so
which is faster to use between the two it terms of data processing?
|
If that is the case then dbf is the one to use. Not because of
speed but of flexibility and ease of use.
No matter what you think the report should print, someone,
somewhere, will want it in a different format, different data,
sorted differently, with different selection criteria.
With a dbf file you just modify the query SQL text.
Remember one thing, hardware gets cheaper all the time,
programming time ever more expensive.
PR |
|
| Back to top |
|
 |
Stephan Eggermont Guest
|
Posted: Thu Aug 03, 2006 7:32 pm Post subject: Re: for flat-file application..which is faster to process da |
|
|
Jamie <jamie_ka1lpa_not_valid_after_ka1lpa_ (AT) charter (DOT) net> wrote:
| Quote: | shotokan99 wrote:
guys i have this scenario:
from dbf files containing around 300k records im going to replicate
these data using txtfile..some sort like of a backup. then im going to
repeat the same process on daily basis.
which is better to use txtfile or dbf file? which is faster?
text files are fine for backing up data of that size but not
for random access.
|
On the other hand, ram is fine for random access while disk is not.
The problem might be simple enough that you can afford to read and
write all data.
Stephan |
|
| Back to top |
|
 |
|