Storage and energy efficient cloud computing.
Authors: Onkar Kulkarni, Kunal Ingale, Rutik kadam, Zaheer shaikh.
Storage and energy efficient cloud computing.
Introduction:
Cloud computing is an emerging technology that attracts ICT(Information and Communication Technology) service providers offering tremendous opportunities for online distribution of services. It offers computing as a utility, sharing resources of scalable data centers. End users can benefit from the convenience of accessing data and services globally, centrally managed backups, high computational capacity, and flexible billing strategies.[2] Suitable, flexible, on-demand service, pooling of resources, broad network access, and measured service are some of the key attrubutes that a cloud can provide with minimal management effort, and highly customizable manner.
The amount of energy consumed by the ICT equipment in data centers need to be reduced. It has been calculated that the amount of energy consumption by the data centers of the service providers is equal to 1.5% of the power supplied to an entire city
A recent calculation figured out 60 kWh to 160 kWh and 160 kWh to 400 kWh energy consumption for the laptop and desktop respectively (if they are on for 2,600 hours per year). Similarly, to cool down a server 200 watts of server power is consumed.
In cloud computing energy conservation is very important because of the worry for the environment and the necessity to decrease greenhouse emissions. To reduce the power requirements some data centers have been built in high altitudes. On the other hand, building a data center at a high altitude also gives rise to several implementation issues. Hence, reducing energy requirements in cloud computing has always attracted the researcher's attention.
Why energy efficient cloud computing is important :
According to amazon.coms’s estimates, data centres expenses related to the cost and operation of the servers account for 53% of the total budget (based on a 3-year amortization schedule), while energy-related costs amount to 42% of the total, and include both direct power consumption (∼19%) and the cooling infrastructure (23%) amortized over a 15-year period. most of data centre power is spent on cooling ICT equipment.
There are several methods to reduce energy consumption in cloud systems.
(i) Energy Efficient Hardware:
‘SpeedStep’ is a technique proposed by Intel and PowerNow technique by AMD, reduce power and heat generation. As the energy consumption is proportional to the use of any resource, this remains as an unsolved problem. So, software-based energy saving schemes was developed. . Novel emerging technologies such as solid-state discs are content with much less energy than the currently used hard disc drives
(ii) Energy-Aware Scheduling:
Different energy-aware methods like Dynamic Voltage Scaling (DVS), Dynamic Voltage/Frequency Scaling (DVFS), Request Batching were developed. But, all these schemes don’t achieve energy optimization significantly.
(iii) Consolidation:
The consolidation technique includes Server Consolidation, Task Consolidation, Energy Task Consolidation (ETC), TaskBased Energy Consumption and Energy Conscious Task Consolidation (ECTC).
(iv) Green Cloud Computing :
Green cloud computing is not only Environment friendly but also energy efficient. Geen Computing introduced to an idea of using renewable energy resources, can make data centers more energy efficient by reusing heat from servers to heat nearby houses, by using recyclable hardware and material which has long lifespan and less toxic material.
Saving on energy consumption will result the overall server management and maintenance costs while still providing best results to the customer.
Architectural Diagram for Green Cloud Computing[1]
(V) Scheduling algorithm :
A scheduling algorithm is the algorithm which dictates how much CPU time is allocate to processes and threads. Algorithms like bin packing, power expand min-max and minimization migrations, highest potential growth algorithms are used to minimize energy utilization cloud.[3]
(VI) Virtual Machine Consolidation :
Virtual machine (VM) consolidation of virtual machines (VMs) to transfer a VM among physical servers, by consolidation we can achieve tremendous saving in power and saving costs. Additional we can reduce the carbon emissions and IT footprint and space as well.
Conclusion :
For cloud computing today, the field of resource management and consumption of energy is an significant and motivating topic. In reality, data centers consume a massive amount of electrical energy that causes a decrease in efficiency and a significant amount of carbon dioxide to be released. Many techniques, such as server virtualization, replication, and consolidation, are used to increase resource utilization and reduce energy usage.
References:
[2] Boru, D., Kliazovich, D., Granelli, F. et al. Energy-efficient data replication in cloud computing datacenters. Cluster Comput 18, 385–402 (2015). https://doi.org/10.1007/s10586-014-0404-x
Comments
Post a Comment