The conclusion of the ETH white paper “An Evaluation of Alternative Architectures for Transaction Processing in the Cloud”

The previous post contains a diagram from the white paper that you can find under:

Here, I would like to post the conclusions:

This paper presented the results of a first study of the end-to-end

performance and cost of running enterprise web applications with

OLTP workloads on alternative cloud services. Since the market is

still immature, the alternative services varied greatly both in cost

and performance. Most services had significant scalability issues.

An interesting observation was to see how the alternative services

behave in overload situations. With regard to cost, it became clear

that the alternative providers have different business models and

target different kinds of applications: Google seems to be more interested

in small applications with light workloads whereas Azure

is currently the most affordable service for medium to large services.

Public clouds are often criticized for a lack of support to

upload large data volumes. This observation could be confirmed.

It is still difficult to upload, say, 1 TB or more of raw data through

the APIs provided by the providers.

The more fundamental question of what is the right data management

architecture for cloud computing could not be answered.

It is still unclear whether the observed results are an artifact of the

level of maturity of the studied services or fundamental to the chosen

architecture. We hope that this work has pathed the way to a

continuous monitoring of progress on alternative approaches and

products for data management in the cloud.

I have to admit, I love the statement: “…whereas Azure is currently the most affordable service for medium to large services.”.