ieee 2014 java cloud computing project dynamic heterogeneity-aware resource provisioning in the...

7/21/2019 IEEE 2014 JAVA CLOUD COMPUTING PROJECT Dynamic Heterogeneity-Aware Resource Provisioning in the Cloud

http://slidepdf.com/reader/full/ieee-2014-java-cloud-computing-project-dynamic-heterogeneity-aware-resource 1/3

Dynamic Heterogeneity-Aware Resource

Provisioning in the Cloud

Abstract

Data centers consume tremendous amounts of energy in terms of power distribution and

cooling. Dynamic capacityprovisioning is a promising approach for reducing energy

consumption by dynamically adjusting the number of active machines tomatch resource

demands. However, despite extensive studies of the problem, existing solutions have not fully

considered theheterogeneity of both workload and machine hardware found in production

environments. In particular, production data centers oftencomprise heterogeneous machines

with different capacities and energy consumption characteristics. Meanwhile, the production

cloudworkloads typically consist of diverse applications with different priorities, performance

and resource requirements. Failure to considerthe heterogeneity of both machines and

workloads will lead to both sub-optimal energy-savings and long scheduling delays, due

toincompatibility between workload requirements and the resources offered by the

provisioned machines. To address this limitation, wepresent Harmony, a Heterogeneity-

Aware dynamic capacity provisioning scheme for cloud data centers. Specifically, we first

use theK-means clustering algorithm to divide workload into distinct task classes with similar

characteristics in terms of resource andperformance requirements. Then we present a

technique that dynamically adjusting the number of machines to minimize total energy

consumption and scheduling delay. Simulations using traces fr om a Google’s compute cluster

demonstrate Harmony can reduceenergy by 28 percent compared to heterogeneity-oblivious

solutions.

Existing Scheme

GLOBALSOFT TECHNOLOGIESIEEE PROJECTS & SOFTWARE DEVELOPMENTS

IEEE FINAL YEAR PROJECTS|IEEE ENGINEERING PROJECTS|IEEE STUDENTS PROJECTS|IEEE

BULK PROJECTS|BE/BTECH/ME/MTECH/MS/MCA PROJECTS|CSE/IT/ECE/EEE PROJECTS

CELL: +91 98495 39085, +91 99662 35788, +91 98495 57908, +91 97014 40401

Visit: www.finalyearprojects.org Mail to:[email protected]

mailto:[email protected]





Despite extensive studies of the problem, existing solutions have not fully considered

theheterogeneity of both workload and machine hardware found in production

environments.it can be easily integratedwith existing scheduling algorithms,variants of first-

fit and best-fit algorithms and Open source platforms such as Eucalyptus canadopt this

mechanism by changing the scheduling policy toweight round-robin first fit and weight

round-robin best fit,respectively. The main benefit of CBP is its simplicity and practicalityfor

deployment in existing systems . The existing work on this topic has notaddressed a key

challenge, which is the heterogeneity ofworkloads and physical machines.

Proposed Scheme

The fact that a large number of DCP schemeshave been proposed in the literature in recent

years, a keychallenge that often has been overlooked or considered difficultto address is

heterogeneity, which is prevalent in production

cloud data centers. present theformulation of the heterogeneity-aware DCP ,

and provide two technical solutions . Discusses the deployment of Harmony in practice.

Finally,we evaluate our proposed system using Google workload

traces. as both scheduler designand capacity upgrade require a careful understanding ofthe

workload characteristics in terms of arrival rate, requirements, and duration .We

haveanalyzed the workload of a Google compute cluster, andproposed an approach to task

classification using k-meansclustering .

Conclusion

Dynamic capacity provisioning has become a promisingsolution for reducing energy

consumption in data centers inrecent years. However, existing work on this topic has

notaddressed a key challenge, which is the heterogeneity of

workloads and physical machines. In this paper, we first providea characterization of both

workload and machine heterogeneityfound in one of Google’s production computeclusters.

Then we present Harmony, a heterogeneity-awareframework that dynamically adjusts the

number of machinesto strike a balance between energy savings and schedulingdelay, while

considering the reconfiguration cost. Throughexperiments using Google workload traces, we



found Harmonyyields large energy savings whilesignificantlyimproving task scheduling

delay

System Configuration:-

Hardware Configuration:-

Processor - Pentium – IV

Speed - 1.1 Ghz

RAM - 256 MB(min)

Hard Disk - 20 GB

Key Board - Standard Windows Keyboard

Mouse - Two or Three Button Mouse

Monitor - SVGA

Software Configuration:-

Operating System : Windows XP

Programming Language : JAVA

Java Version : JDK 1.6 & above.

ieee 2014 java cloud computing project dynamic heterogeneity-aware resource provisioning in the...

Documents