| | |
Summary: RACNet: A High-Fidelity Data Center Sensing Network
Chieh-Jan Mike Liang, Jie Liu, Liqian Luo , Andreas Terzis, Feng Zhao
Johns Hopkins University Microsoft Research Google
{cliang4, terzis}@cs.jhu.edu {jie.liu, zhao}@microsoft.com liqianluo@gmail.com
Abstract
RACNet is a sensor network for monitoring a data cen-
ter's environmental conditions. The high spatial and tempo-
ral fidelity measurements that RACNet provides can be used
to improve the data center's safety and energy efficiency.
RACNet overcomes the network's large scale and density
and the data center's harsh RF environment to achieve data
yields of 99% or higher over a wide range of network sizes
and sampling frequencies. It does so through a novel Wire-
less Reliable Acquisition Protocol (WRAP). WRAP decou-
ples topology control from data collection and implements a
token passing mechanism to provide network-wide arbitra-
tion. This congestion avoidance philosophy is conceptually
different from existing congestion control algorithms that
retroactively respond to congestion. Furthermore, WRAP
adaptively distributes nodes among multiple frequency chan-
|
