big o notation for reading data in blocks

Unlikely, but it may be conceivable that some implementation could have been written to behave that way.

I did a program that read the first o characters and then the second o characters. But, I am not sure of the complexity. To read the entire file, you would use p/o blocks. So, the complexity is (p/o)*p. However, I am not sure if this is correct for this implementation. Thanks a lot.

Thanks for replying.

I only read a block each time (and not the entire file).

So what is the time to read the first o characters?

The time to read the first o characters is O(o).
The time to read the second o characters is O(o)
The time to read the last o characters is O(o).

At first, I thought, it should be added. O(o)+O(o) which gives O(p). But, sometimes, I see people put the number of blocks or the number of rounds that is required to read the entire O(p). So, p/o gives you the number of blocks or rounds. So, you need p/o rounds or blocks to read the entire file. I see, you don't times with p but with o. That is, (p/o)*o. So, it is O(p) I think.

I understand that constant is ignored. p/o looks like a constant but it would given different values for different data. So, it can be considered a variable. That is the reason, I am multipying with  p/o...

That is, (p/o)*o. So, it is O(p/o^2)

Look more carefully.
[Edit: sorry, I see you already did]

Sorry ozo, that is what I wrote and then, I realise that it is o^2.  

This means that the big o notation is very simple. That is, O(p). It doesn't matter whether you read the files in blocks or at once.

Sorry for sounding stupid. p/o is a variable (and not a constant) because it gives a different value for different files. But, it would give similar value for the same input file. Did I understand this wrong ?

What if you try using actual values for demonstration?

Let's say you have a file of length 10 bytes (p). If you read 1 byte at a time, then you have 10 rounds, and that would be O(p). If you now break that up into 2 byte chunks (o), you now have 5 rounds. Since it takes 5 rounds to read the file, should the overall time not be O(p/o)--the overall size of the file divided by the size of the chunk. (I am assuming you're considering the time taken to read a chunk to be constant time, and not factoring in the reading of each byte into the complexity of reading a chunk.) I am not sure why you would want to multiply p or o to the division.

Hi kaufmed,

I am a newbie in big o notations (I am trying to understand myself). This is what I understand (I may be wrong).

Say o is 5 and p is 10
It takes O(5) to read the first 5 characters
Another O(5) to read the second 5 characters

In total, it takes O(5+5) = O(10)
Now, we can get the same number if take the number of rounds which is 2 here and multply it with O(5) which gives O(10).

The confusion is that you need to ignore the constants but you keep the variables..

Thanks very much for the replies..

I think, i understand now. When p doubles, the time doubles but the doubling is never reflected in the big oh notation.

I would say that the doubling is reflected in the big oh notation in the sense that
O(p) implies that when p doubles, the time doubles