Journal of Theoretical and Applied Information Technology
31
st
December 2019. Vol.97. No 24
© 2005 – ongoing JATIT & LLS
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195
3730
ENERGY EFFICIENCY AND SCHEDULING TECHNIQUES
USING CLOUD COMPUTING METHODS WITHIN DATA
CENTERS
YAO FRANCOIS MICHAEL KRA
1
, SAAD MASOOD BUTT
2,
NOAH KWAKU BAAH
3
, IMRAN
MEMON
4
, BOAFOH KYEI BAFFOUR
-5
, FRANCK ADDO MANTE-
6
, MALO TEHINKE
ACHILLE
-7
1- 3
School of Computer Science and Engineering, Southeast University, Nanjing, Jiangsu Province, China
2-
Atlantic International University, Honolulu, HI 96813, United States
4-
College of Computer Science, Zhejiang University, Hangzhou, Zhejiang Province, China
5-6
College of Computer and Information Engineering, Hohai University, Nanjing, Jiangsu Province, China
7-
Institute of Functional Nano and Soft Materials, Soochow University, Suzhou, Jiangsu Province, China
ABSTRACT
As knowledge increases and technology advances, a large pool of data has become available to man. With
data growing endlessly every day, data centers have become important in ICT, emphasizing the need to
ensure efficient use of energy. Due to the high quantity of electricity consumed by these data centers, Data
centers in the U. S. alone has been estimated to consume about 92-billion -kilowatt-hours of voltage during
the year 2013. By 2020, the power consumed is predicted to reach 140-billion kilowatt hours and estimated
to causes release of about 145-million metric-tons of carbon substances yearly. Datacenters have rapidly
come to be the pillar of the global mobile system. In the present paper, we surveyed on various techniques
used for efficient scheduling of tasks in datacenters and also reviewed the traditional scheduling algorithms
advantages and their disadvantages to better insight new researchers in this field of research. We discussed
contributing issues to data centers energy inefficiency such as Physical servers, Energy-efficient hardware,
energy efficient workflows, Cooling Equipment, and Power plants. We recommended some optimal
techniques and prospect directions regarding the future development of new algorithms to overcome some
of the energy inefficiency drawbacks in the cloud data centers.
Keywords: Data Green Centers, Task scheduling, Cloud Computing, Energy Efficiency, Scheduling
Algorithms
1. INTRODUCTION
Resourcefully big data centers cause a significant
problem Data Center’s consumption of power has
substantially augmented for decades to an
extraordinary level. Today’s ICT economy growth,
incentives, and data digitization have based on cloud
platforms. Organizations run huge data centers and
transfer large data amongst their sub-branches
Infrastructure providers. For that matter, power
consumption in the United Stated of America where
most of the largest data centers such as Google,
Facebook, Amazon, IBM, and others has located is
in prominently confronted with an increment of
expenditure growth for data centers’ electricity
usage. Based on a report by National Resources
Defense Council stated that, “In 2013, the U.S. Data
Centre consumed an estimated average 91 billion
kilowatts per hours of electricity”. It has shown that
electricity used supply all the inhabitants in New-
York-City keep doubling for the past years.
Moreover, is also on track to increase and submerge
an estimated average of 140 billion kilowatts per
hours by the year 2020. The equivalency which has
been the annual output of thirty-four (34) massive
“500-megawatt coal-fired power plants”. The
electricity usage in Data center is forecasted to rise
to about “140 billion kilowatt-hours annually by
2020”, which estimated value would be
correspondent to the yearly output of about 49
energy plants. The costs which arise cause American
businesses about “$13 billion annually in electricity
bills and emits close to 100 million metric tons of
carbon pollution per year”. Power demands for
computing have been expanding steadily due to