Seek Time vs Transfer Rate

Asked

Viewed 236 times

1

I was reading a bit about the Hadoop tool for big data and came across the terms Seek Time and Transfer Rate. I found definitions like Seek Time is the delay to find a specific file/data and Transfer Rate is the speed to find the file/data but I’m not very satisfied with the answers I found and I’m not sure they are right.

So I’d like to know if anyone has more detailed definitions of the two terms and could answer those questions.

  1. What is Seek Time and Transfer Rate?

  2. What’s the real difference between them ?

  3. And when it’s best to wear one and when it’s best to use the other ?

hadoop

  • in this case the most important and the tranfer, where the Mapreduce is based on the speed to the data.

1 answer

0

Seek Time is the delay (the time it takes) for the system to find the file (place the read/write head in the file).

Transfer Rate, in the context of throughtput, is the read or write speed of that file.

Both are useful and measure different things.

RDBMS being based on B-Tree, is highly dependent on Seek operations and is therefore more efficient to update small portions in large data sets.

Mapreduce uses SORT and Merge operations to recreate data, which ends up being more dependent on Transfer.

As Transfer is faster than Seek and has evolved due to the emergence of new technologies, Mapreduce is best suited for use in Bigdata scenarios.

Browser other questions tagged hadoop

You are not signed in. Login or sign up in order to post.