Qvd size greater than Excel sheet

[nikhilgarg]

Hey,

I ahve an excel sheet of size: 15 KB.

When i convert it to QVd, size : 19 KB

It is being said that QVD's compresses data but here size increases in qvd.

Can anyone explain the reason?

Thanks.

[jonathandienst]

QVD's do not store compressed data - they store the data in a compact form, but not compressed. Excel xlsx format files are compressed, so it should not be unexpected that the XLSX file is smaller than the QVD file.

[tresesco]

QVD stores data with addtional headers. When the data size is very small like yours, it might eventually show the size bigger than the source. But with increasing size of data the header remains same (if the field no. doesn't increase) or increase comparitively with minor size, you can see the difference.

And qvd data is not really compressed data, rather it's duplication-removed data. So you can see the effect if you try same value(say a number-100) to generate 10000 times and store in a qvd, and a sequential series of number (say 1-10000) and store; the second qvd would be bigger than the first one, because of several unique values.

[nikhilgarg]

Hey,

Thanks but we usually say we should create qvd's and fetch data from qvd's bcoz it stores data in compresses form and has high loading speed. So,

1: Isn't qvd's store data in compressed form?

2: Also we use qvd's for loading speed or for minimum size of data?

Thanks

[nikhilgarg]

Hey,

Thanks but we usually say we should create qvd's and fetch data from qvd's bcoz it stores data in compresses form and has high loading speed. So,

1: Isn't qvd's store data in compressed form?

2: Also we use qvd's for loading speed or for minimum size of data?

[jonathandienst]

1. No

2. The QVD format is designed for fast loading into QVW documents and if you have an optmised load, it is several times faster than loading from an equivalent text file or Excel file.

[jonathandienst]

tresesco wrote:

And qvd data is not really compressed data, rather it's duplication-removed data. So you can see the effect if you try same data(say a number-100) to generate 10000 times and store in a qvd, and a sequential series of number (say 1-10000) and store; the second qvd would be bigger than the first one, because of several unique values.

That's part of the picture - the symbol table. The QVD also stores the table as bit-stuffed pointers.

Symbol Tables and Bit-Stuffed Pointers

[nikhilgarg]

Ohk,

So with your reply, i assume that:

1: QVD's does not store the data in compressed form.

2: QVD's are used because loading data from QVD's is always faster than any other source.

But can you please explain how QVD's loading is faster than other data source when we have same data in both and also data in QVD's is not in compressed form?