Mobile networks stand to benefit the most from energy efficiency
efforts, as they are the most inefficient and yet the fastest growing
networks in terms of data volumes. Mobile networks could realize potential energy efficiency improvements of up to 1043 times.
efficiencies in fixed-line and core networks are also expected, but
will be less dramatic. The modeling shows potential improvements in
fixed access networks of 449 times and improvements in the core network
of 95 times. Such networks are already relatively energy-efficient, so
further gains will be less significant and much harder to achieve than
with mobile networks.
Among the technologies that should be
brought in to achieve improved energy efficiency are small-cell
basestations for dense urban environments, shared infrastructure across
operators, use of sleep-modes in the network for periods without
traffic, dynamic allocation of resources, beam-forming of signal
propagation in cells using antenna technology.
Klein said the
GreenTouch-developed BI-PON protocol is demonstrating a 10 to 30 time
energy-saving in networks depending on how aggressively sleep modes are
used. With regard to beam-forming he said that the calculations are
non-trivial because the extra energy cost of doing complex DSP
calculations has to be weighed against the RF energy saved.
want to create a road-map for equipment makers and let people know this
is achievable. There are more technologies and more projects to come
that have not been taken into account," said Klein.
will continue its work on network architectures and technologies with
updates to Green Meter calculations expected later in 2013 and
demonstrations of key technologies planned for 2014 and 2015.
Click on image to enlarge.
GreenTouch membership as of May 2013. Source: GreenTouch
The difference between wishful thinking and reality is a plan with evidence. How much energy efficiency will be gained through efficient use of idle equipment? How much will be achieved by migrating from inefficient to existing efficient technology? How much requires development of new technologies? This article describes a great dream; how will it be realized?
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.