The Glasgow Raspberry Pi Cloud: A Scale Model for Cloud

The Glasgow Raspberry Pi Cloud:
A Scale Model for Cloud
Computing Infrastructures
By: Mugasa Hatwib
Cloud Data Center
• Most research in Data Centers using Cloud
computing is by uses Development
Environments with a handful of machines.
• However this is very expensive for most
research and educational institutions
considering the data center space, power,
cooling, etc.
Google Cloud Data Center
Google Data Center
Cooling system at Google
Credit :
Cloud DC Simulator
• Are a low cost alternative to the development
• However some essential Cloud Computing
properties that these simulators have failed to
capture, e.g. Traffic patterns, cross layer
correlation between application and network.
• Cloud simulation tools also require physical
Cloud Dev environments running on x86
processors which are very expensive.
PiCloud : A scaled Cloud DC model
• The low-power, low-cost Raspberry Pi
provides an affordable option to construct a
miniature Cloud Data Center.
• It also allows for the reproduction of actual
traffic patterns with realistic cloud
System Design
56 x Raspberry Pi Model B
ARM - 700 MHz processor
1 Ethernet Port 10/100
16GB Sandisk SD card
Credit :
System Design
Credit :
System Design
• Built using 56 Model B version Raspberry Pi
devices housed in a rack constructed from
Lego bricks
• Devices in the same rack are connected to the
same Top Of rack switch
System Architectural Design
Credit :
System Virtualization
• Linux LXC Containers were used for virtualization
• API daemon on each Pi providing a RESTful
management interface
• A webserver on the pimaster was used to provide
a web-based control panel for users and
• The website is used to control workloads using
RESTful interfaces through the interaction with
API daemons
System Virtualization
Credit :
• Possible to experiment with new algorithms for
Virtual Machine (VM) management at low cost.
• Directly observing resulting behaviors on all
layers of the Cloud architecture unlike with
• Observations on the interaction and effects of
specific optimizations on the cloud configurations
can be made at all system layers.
• Can be used to investigate ways of reducing
network congestion through improved
resource allocation,
• Can be used to look at novel network
architectures and technologies that require
significant changes to the infrastructure.
• A future development in the Cloud computing
is the adjustment or removal of virtualization
• [1] Fung P.T., White D.R., Jouet S., Singer J., Pezaros D.P., "The
Glasgow Raspberry Pi Cloud: A Scale Model for Cloud Computing
Infrastructures," icdcsw, pp.108-112, 2013 IEEE 33rd International
Conference on Distributed Computing Systems Workshops
(ICDCSW), 2013
• [2] A. Greenberg, J. Hamilton, D. A. Maltz, and P. Patel, “The cost of
a cloud: Research problems in data center networks,” ACM
SIGCOMM Computer Communication Review, vol. 39, no. 1, January
• [3] “Raspbian.” [Online]. Available: (4/29/2014)
• [4] “Raspberry pi.” [Online]. Available:

similar documents