Muli Ben-Yehuda, M.Sc. Thesis Seminar
Wednesday, 1.4.2015, 14:30
In the near future, cloud providers will sell their users virtual
machines with CPU, memory, network, and storage resources whose prices
constantly change according to market-driven supply and demand
conditions. Running traditional operating systems in these virtual
machines is a poor fit: traditional operating systems are not aware of
changing resource prices and their sole aim is to maximize performance
with no consideration for costs. Consequently, they yield low profits.
We present nom, a profit-maximizing operating system designed for
cloud computing platforms with dynamic resource prices. Applications
running on nom aim to maximize their profits from their resources by
optimizing for both performance and costs. The nom kernel provides
them with direct access to the underlying hardware and full control
over their private software stacks. Since nom applications know there
is no single ``best'' software stack, they adapt their stacks'
behavior on the fly according to the current price of available
resources and their private valuations of them. We show that in
addition to achieving up to 3.9x better throughput and up to 9.1x
better latency, nom applications yield up to 11.1x higher profits when
compared with the same applications running on Linux and OSv.