Load Balancing in Cloud Environment

Load Bala i g i Cloud
E iro e t
C“I

: Cloud Co puti g

P ese ted B :
Rash i Pethe
A uta “a eka

Date:

/ 9/

Age da
. I t odu tio
. Ba kg ou d
. Challe ges
. Load Bala i g App oa hes
. Pe fo


a e E aluatio

. Co lusio a d Futu e Wo k

I trodu tio
● U ifo

dist i utio of o kloads

a d o puti g esou es.
● Load i ludes: Net o k Load,
Me o

Load, Dela Load, et .

● Resou es a

e: VM’s, Ha d a e,

Ba d idth et .

● P o ides effe ti e esou e utilizatio
a d

i i al espo se ti e.

I porta e of Load Bala i g
● Why is load ala i g i porta t?
○ Poo utilizatio of esou es
○ Pe fo

a e deg adatio

● Goals of Load Bala i g
○ Li ited e e g

o su ptio

○ Redu i g a o e issio
○ Mi i izi g the jo e e utio ti e a d aiti g ti e i


ueue

Classifi atio of LB Algorith s
● Based o

ho i itiated the pro ess:

○ “e de -i itiated
○ Re ei e -i itiated
○ “
● Based o

et i
urre t State of the syste

○ “tati Algo ith
■ Rou d-Ro i , A ti e-Mo ito i g, et .
○ D a i Algo ith
■ Th ottled, Heu isti Algo ith s, et .


Challe ges
● Auto ated “e i e P o isio i g
○ Use o elease esou e ithout pe fo a e deg adatio
● Vi tual Ma hi es Mig atio s
○ Dist i ute o kload hile VM ig atio
● E e g Ma age e t
○ E e g effi ie t use of esou es
● “to ed Data Ma age e t
○ Dist i utio of data fo opti u sto age a d ith fast a ess
● E e ge e of s all data e te s fo loud o puti g.
○ Load ala i g at glo al s ale to e su e a ade uate espo se

Approa hes of LB Algorith s
● Rou d Ro i LB Algorith :
○ Basi “tati te h i ue.
○ “i ple to i ple e t a d faste
e uest p o essi g.
○ Mai tai s list of VM’s.
○ Da


a k: Bu st ti e of e uest ot

o side ed.
○ Modified e sio s a aila le.

Approa hes of LB Algorith s Co t
● Throttled LB Algorith :
○ Basi D a i te h i ue.
○ Mai tai s allo atio ta le of all VM’s
alo g ith thei states.
○ A aila le: VM id
○ Da
fo

a k: E e

Bus : ti e the sea h egi s

fi st i de .


○ Modified algo ith s a aila le to i p o e
espo se ti e.

Approa hes of LB Algorith s Co t
● Opti al LB Algorith :
○ I p o e e t o e a ti e- o ito i g
algo ith .
○ A ti e Mo ito i g algo ith

usuall looks

fo the least loaded VM.
○ Fe

esou es

a d fe

a


e al a s e us

ight e u de utilized.

○ I p o es pe fo

a e of the s ste s

hile edu i g the espo se ti e.

Approa hes of LB Algorith s Co t
● Mi -Mi LB Algorith :
○ T aditio al
○ Mi i u
i tual

i - i is si ple a d Fast.
o pletio ti e fo ea h task is o puted fo all

a hi es.


○ The task ith o e all

i i u

○ Re uest es heduli g is pe fo

o pletio ti e is hose .
ed to use u de utilized

esou es.
○ I p o es o e all
○ Effe ti e he

akespa .
u

e of s alle tasks e eeds la ge o es.

source:https://ac.els-cdn.com/S1877050915019146/1-s2.0-S1877050915019146-main.pdf?_tid=c6a8bd54-d2ea-11e7-b23c-00000aacb361&acdnat=1511729325_7e357e774605b2f82ac88e14e342ced1


Perfor a e E aluatio Metri s

● Metri s for LB Algorith s
○ Th oughput
○ Asso iated o e head
○ Fault tole a e
○ Mig atio ti e
○ Respo se ti e
○ Resou e utilizatio
○ “ ala ilit
○ Pe fo

a e

E aluatio

ith CloudA alyst

● Tool fo e aluatio a d

odeli g of Cloud Apps
● Co figu e pa a ete s fo
si ulatio i GUI
● Basi “etti gs
○ No. of DC’s ○ No. of UB’s ○ “e i e oke poli Closest DC
○ Algo ith s -

E aluatio Metri - Respo se Ti e
● Rou d -Ro i a d A ti e Mo ito i g ha e si ila a e age espo se
ti es.
● Th ottled algo ith

has the ette esults.

E aluatio Metri - Pro essi g Ti e
● P o essi g ti e take
e ause of d a i

th ottled is less o pa ed to othe s,
atu e of this algo ith .


● Rou d-Ro i a d A ti e-Mo ito i g algo ith
ti es.

ha e si ila p o essi g

Si ulatio Results Vie

Co lusio & Future Work

● Challe gi g to s hedule o kload u ifo l i dist i uted
e i o e t.
● Possi ilit of ha i g u de loaded a d o e loaded esou es
i eases he o kload is dist i uted a do l .
● Dis ussed fou asi load ala i g app oa hes a d e aluated
ith Cloud-A al st tool.
● Th ottled algo ith is d a i so ette suit a o gst all fou
algo ith s dis ussed.
● Futu e o k i ol es i ple e ti g a d si ulati g Mi -Mi a d
Opti al Load Bala i g i Cloud A al st.

Refere es
.
.
.
.
.
.
.

Do a al, “h idha G., a d G. Ra Moha a Redd , Opti al load ala i g i loud o puti g
effi ie t utilizatio of i tual a hi es , Co
u i atio “ ste s a d Net o ks COM“NET“ , “i th
I te atio al Co fe e e o IEEE,
.
“ha , “u had a Bose, a d A. K. “i gh, A su e o s heduli g a d load ala i g te h i ues i loud
o puti g e i o e t , Co pute a d Co
u i atio Te h olog
ICCCT , I te atio al
Co fe e e o . IEEE,
.
Al Nuai i, Klaithe , A su e of load ala i g i loud o puti g: Challe ges a d algo ith s ,
Net o k Cloud Co puti g a d Appli atio s NCCA , “e o d “ posiu o IEEE,
.
Che , Hua kai, Use -p io it guided Mi -Mi s heduli g algo ith fo load ala i g i
loud
o puti g , Pa allel Co puti g Te h ologies PARCOMPTECH ,
Natio al Co fe e e o IEEE,
.
Ga g, “hikha, D. V. Gupta, a d Rakesh Ku a D i edi. E ha ed A ti e Mo ito i g Load Bala i g
algo ith fo Vi tual Ma hi es i loud o puti g. “ ste Modeli g & Ad a e e t i Resea h
T e ds “MART , I te atio al Co fe e e. IEEE,
.
Nus at Pasha, D . A it Aga al, D . Ra i Rastogi. Rou d Ro i App oa h fo VM Load Bala i g
Algo ith i Cloud Co puti g E i o e t I te atio al Jou al of Ad a ed Resea h i Co pute
“ ie e a d “oft a e E gi ee i g,Volu e , Issue , Ma
.
“tuti Da e, P asha t Maheta. Utilizi g Rou d Ro i Co ept fo Load Bala i g Algo ith at Vi tual
Ma hi e Le el i Cloud E i o e t I te atio al Jou al of Co pute Appli atio s, Volu e 9 -No
, Ma
.

THANK YOU
QUE“TION“

Dokumen yang terkait

Analisis Komparasi Internet Financial Local Government Reporting Pada Website Resmi Kabupaten dan Kota di Jawa Timur The Comparison Analysis of Internet Financial Local Government Reporting on Official Website of Regency and City in East Java

19 819 7

ANTARA IDEALISME DAN KENYATAAN: KEBIJAKAN PENDIDIKAN TIONGHOA PERANAKAN DI SURABAYA PADA MASA PENDUDUKAN JEPANG TAHUN 1942-1945 Between Idealism and Reality: Education Policy of Chinese in Surabaya in the Japanese Era at 1942-1945)

1 29 9

Implementasi Prinsip-Prinsip Good Corporate Governance pada PT. Mitra Tani Dua Tujuh (The Implementation of the Principles of Good Coporate Governance in Mitra Tani Dua Tujuh_

0 45 8

Improving the Eighth Year Students' Tense Achievement and Active Participation by Giving Positive Reinforcement at SMPN 1 Silo in the 2013/2014 Academic Year

7 202 3

Improving the VIII-B Students' listening comprehension ability through note taking and partial dictation techniques at SMPN 3 Jember in the 2006/2007 Academic Year -

0 63 87

An Analysis of illocutionary acts in Sherlock Holmes movie

27 148 96

The Effectiveness of Computer-Assisted Language Learning in Teaching Past Tense to the Tenth Grade Students of SMAN 5 Tangerang Selatan

4 116 138

The correlation between listening skill and pronunciation accuracy : a case study in the firt year of smk vocation higt school pupita bangsa ciputat school year 2005-2006

9 128 37

Existentialism of Jack in David Fincher’s Fight Club Film

5 71 55

Phase response analysis during in vivo l 001

2 30 2