After calculating the average spot instance price, I made an overview of the spot, on-demand and reserved instance prices and this for all different geographical regions. Also the prices for storage in these regions are included in the overview, which can be found here. We notice that for all instances except for a standard large linux instance the US East region is the cheapest one. For storage only the US West region is a bit more expensive than the other regions. For data transfer on the other hand, the Asia region is the most expensive one. For every region I made a graph (they can be found here) that plots all the different prices, an example is given here:
Afterwards I had a look at the optimal division between reserved and spot instances, but without taking into account more properties of using spot instances (such as the need for checkpointing, ...). Using spot instances seemed to be cheaper than using reserved instances, such that the optimal division is using a 100 percent spot instances.
During the thesis meeting last week another question was raised: whether it is cheaper to under- or overestimate the number of required Reserved instances to reach an optimal division? While trying some different workloads in the earlier created Excel sheet, I found out this (as one would expect) depends on the workload. I tried to find a formula that expressed whether for a certain workload it would be better to take one less or one more reserved instance. Of course this is a bit contradictious, since you could determine the optimal amount of reserved instances if the workload is known. But maybe the found tipping point could tell us something, my findings can be found here.
Monday, October 18, 2010
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment