>  Hi,

>  I've looked at protocol buffers, and I've noted that there is no
> support for arrays
>  of values (double, integers). This is a significant drawback, for
> example
>  JSOM, HDF5 etc they all have this.

>  One post suggested  that one should put an array  as one single
> string in a field
>  I've did this, and the performance  was very bad in Java and very
> memory consuming
>  (compare to the standard Java serialization).
>  I've wrote 500 times the same array (10,000 double numbers ), and
> after the array 500,
>  my computer was out of memory,

>  Secondly, all tutorials suggest that the file should be written at
> once, i.e. at the end
>  of the program, when the messages
>  are filled. I want to write data to the disk in several steps. say I
> want to write one record first (say, one array),
>  then I want to append data to the existing file, and so on, this way
> I will not need
>  to keep all records in the computer memory. The merge mechanism
> shown in the tutorial
>  seems parses the old file
>  first, and then add new record, and write a new file.

>  Do I understand this correctly? If yes, then the protocol  buffers is
> not too good for large data volumes,
>  especially with numerical arrays

>  best wishes, Sergei