Hi All ,
I need to moving average using cascading.
in case there is data input and moving average window size is 3 I need to take averages of :
1 2 3 5 6 3 2 5 7 3 6 5
so in case of widow size 3 the buckets which I need to calculate are the following:
(1 2 3) , (2 3 5) , (3 5 6) , (5 6 3) , (6 3 2) , (3 2 5) , (2 5 7) , (5 7 3) , (7 3 6) , (3 6 5)
I plan to use cascading Aggregator ( or Buffer ) with RingBuffer.
My question is what is the internal implementation in execution time?
Suppose I have huge amount of data , but in order to calculate moving averages I need to have only 3 values at the same time. Does Buffer / Aggregator provide ability to process data using small buckets or it will required to hold all data in memory?
Thanks
Oleg.