Fast forward-backward (write-read)

[Virgil Stokes]

On 23-Oct-2012 18:09, Tim Chase wrote:
> On 10/23/12 09:31, Virgil Stokes wrote:
>> I am working with some rather large data files (>100GB) that contain time series
>> data. The data (t_k,y(t_k)), k = 0,1,...,N are stored in ASCII format. I perform
>> various types of processing on these data (e.g. moving median, moving average,
>> and Kalman-filter, Kalman-smoother) in a sequential manner and only a small
>> number of these data need be stored in RAM when being processed. When performing
>> Kalman-filtering (forward in time pass, k = 0,1,...,N) I need to save to an
>> external file several variables (e.g. 11*32 bytes) for each (t_k, y(t_k)). These
>> are inputs to the Kalman-smoother (backward in time pass, k = N,N-1,...,0).
>> Thus, I will need to input these variables saved to an external file from the
>> forward pass, in reverse order --- from last written to first written.
>>
>> Finally, to my question --- What is a fast way to write these variables to an
>> external file and then read them in backwards?
> Am I missing something, or would the fairly-standard "tac" utility
> do the reversal you want?  It should[*] be optimized to handle
> on-disk files in a smart manner.
Not sure about "tac" --- could you provide more details on this and/or a simple 
example of how it could be used for fast reversed "reading" of a data file?
>
> Otherwise, if you can pad the record-lengths so they're all the
> same, and you know the total number of records, you can seek to
> Total-(RecSize*OneBasedOffset) and write the record,optionally
> padding if you need/can.  At least on *nix-like OSes, you can seek
> into a sparse-file with no problems (untested on Win32).
The records lengths will all be the same and yes seek could be used; but, I was 
hoping for a faster method.

Thanks Tim! :-)
>
> -tkc
>
>
>
> [*]
> Just guessing here. Would be disappointed if it *wasn't*.