A Correlation-based Methodology to Infer Communication patterns - - PowerPoint PPT Presentation

SLIDE 1

A Correlation-based Methodology to Infer Communication patterns between Cloud Virtual Machines Claudia Canali Riccardo Lancellotti

Dept of Engineering “Enzo Ferrari” University of Modena and Reggio Emilia

SLIDE 2

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 2

M

• t

i v a t i

• n
• T

h e c h a l l e n g e s

• f

e n e r g y e ffjc i e n c y i n D a t a C e n t e r s

– M

u l t i p l e H e t e r

• g

e n e

• u

s V M s

– M

u l t i p l e R e s

• u

r c e s ( C P U , M e m

• r

y , N e t w

• r

k i n g )

• T

h e c h a l l e n g e s

• f

C l

• u

d C

• mp

u t i n g

– D

y n a m i c e n v i r

• n

m e n t

– C

• m

p l e x S L A t

• m

e e t

+ =

SLIDE 3

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 3

M

• t

i v a t i

• n
• T

h e c h a l l e n g e s

• f

e n e r g y e ffjc i e n c y i n D a t a C e n t e r s

– M

u l t i p l e H e t e r

• g

e n e

• u

s V M s

– M

u l t i p l e R e s

• u

r c e s ( C P U , M e m

• r

y , N e t w

• r

k i n g )

• T

h e c h a l l e n g e s

• f

C l

• u

d C

• mp

u t i n g

– D

y n a m i c e n v i r

• n

m e n t

– C

• m

p l e x S L A t

• m

e e t

• →

D a t a c e n t e r ma n a g e me n t i s a t

• u

g h g a me !

+ =

SLIDE 4

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 4

T h e c r i t i c a l r

• l

e

• f

n e t w

• r

k i n g

• T

y p i c a l l y n

• t

c

• n

s i d e r e d i n e x i s t i n g e n e r g y mo d e l s

• I

n t e r a c t i

• n

a mo n g V M s

• I

mp a c t

• f

n e t w

• r

k p a t t e r n s

• n

:

– P

e r f

• r

ma n c e : S L A s a t i s f a c t i

• n

a fg e c t e d b y l a t e n c y

– E

n e r g y : N e t w

• r

k i n f r a s t r u c t u r e c

• n

s u m e s a n

• n
• n

e g l i g i b l e a m

• u

n t

• f

e n e r g y

• E

v

• l

u t i

• n

t r e n d :

– N

e t w

• r

k i m p

• r

t a n c e i s c r i t i c a l

– N

e t w

• r

k i n g i s g

• i

n g v i r t u a l : V R , N O S , S D N

– →

I n t r

• d

u c i n g t h e S D D C

SLIDE 5

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 5

I n t r

• d

u c i n g S D D C

VM Mon Mon Mon VM VM VM Host Host Host Net device Net device Mon Mon SLA VM Mgr Net Mgr 3 1 2 1 1 1 1 2 3 3 3 3 Virtual Comp Net Managment 1: VM placement, VM VM VM VM VM VM VM R R Host Host Host SDN Device SDN Device

Net Mon Net Mon Net Mon VM Mon VM Mon VM Mon

SLA

Net Mon Net Mon VR Mon VR Mon

1 2 3

Comp model Net model Migration model Time model

SDDC Manager

Comp data Net data

1 1 1 1 1 1 1 1 2 3 3 Managment Virtual Comp & Net 1: VM placement, VM

SLIDE 6

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 6

K n

• w

i n g n e t w

• r

k p a t t e r n s

• M

a n a g e me n t i n S D D C r e q u i r e s k n

• w

l e d g e

• f

n e t w

• r

k p a t t e r n s

– Wh

i c h V M s e x c h a n g e d a t a ?

– A

v a i l a b l e i n f

• r

m a t i

• n

: → A g g r e g a t e d a t a

• H
• r

i z

• n

t a l r e p l i c a t i

• n

:

– M

u l t i p l e V M s h a v e s i m i l a r n e t w

• r

k p a t t e r n s

• →

O p e n c h a l l e n g e t

• a

d d r e s s

SLIDE 7

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 7

G

• a

l

• I

n p u t :

– T

r a ffj c p a t t e r n

• f

e a c h V M

– T

i m e s e r i e s

• f

p k t i n /

• u

t

• O

u t p u t :

– V

M s i n t e r a c t i

• n

m a t r i x

• C

a v e a t s :

– P

r e s e n c e

• f

h

• r

i z

• n

t a l r e p l i c a t i

• n

– D

a t a s a m p l e s m a y b e n

• t

s y n c h r

• n

i z e d

VM1 VM2 VM3 VM4 VM5 VM6 VM7 VM8 VM9 VM1 1 1 1 VM2 1 1 1 VM3 1 1 1 VM4 1 1 1 VM5 1 1 1 VM6 1 1 1 VM7 1 1 1 VM8 1 1 1 VM9 1 1 1

SLIDE 8

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 8

M e t h

• d
• l
• g

y

• S

y n c h r

• n

i z a t i

• n
• f

t i me s e r i e s

– C

u b i c i n t e r p

• l

a t i

• n
• f

s a m p l e s

• C
• mp

u t a t i

• n
• f

c

• r

r e l a t i

• n

ma t r i x

– C

• m

p u t e s c

• r

r e l a t i

• n

m a t r i x b e t w e e n a l l t h e ( s y n c h r

• n

i z e d ) t i m e s e r i e s

– M

u l t i p l e c

• r

r e l a t i

• n

i n d e x e s a r e c

• n

s i d e r e d

• I

d e n t i fj c a t i

• n
• f

i n t e r a c t i n g V M s

– U

s e

• f

t h r e s h

• l

d

– M

• r

e c

• m

p l e x a p p r

• a

c h e s m a y b e u s e d

Sync. Correlation Threshold.

SLIDE 9

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 9

C

• r

r e l a t i

• n

i n d e x e s

• P

e a r s

• n

c

• r

r e l a t i

• n

c

• e

ffjc i e n t

• S

p e a r ma n c

• r

r e l a t i

• n

c

• e

ffjc i e n t b a s i c a l l y w e a p p l y t h e P e a r s

• n

c

• r

r e l a t i

• n

t

• t

h e t i m e s e r i e s

• f

r a n k s f

• r

e a c h v a l u e i n t h e

• r

i g i n a l s a m p l e s .

• S

p e a r ma n t e n d s t

• a

mp l i f y s ma l l

• s

c i l l a t i

• n

s a r

• u

n d a v e r a g e v a l u e

SLIDE 10

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 1

E x p e r i me n t a l s e t u p

• E

x p e r i me n t s

• n

A ma z

• n

E C 2

– U

s e

• f

m i c r

• i

n s t a n c e s

• T

h r e e

• t

i e r We b a p p l i c a t i

• n

b e n c h ma r k : T P C

• W

– 4

v e r t i c a l s t a c k s , 3 V M s p e r s t a c k

• D

a t a c

• l

l e c t i

• n

i n t e r v a l :

– 3

s e c , 1 m i n , 2 m i n

• M

e t r i c s

• f

i n t e r e s t

– P

r e c i s i

• n

( T P / T F + T P )

– R

e c a l l ( T P / T P + F N )

– A

c c u r a c y ( T P + T N / T P + T N + F P + F N )

SLIDE 11

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 1 1

Q u a l i t a t i v e a n a l y s i s

• U

s e

• f

h e a t ma p

• I

d e a l r e s u l t :

– R

e d b

• x

e s

• n

d i a g

• n

a l

– B

l u e e v e r y w h e r e e l s e

• P

e a r s

• n

c

• e

ffjc i e n t

– C

• r

r e l a t i

• n

a l w a y s h i g h

– L

a r g e r e d h a l

• s
• S

p e a r ma n c

• e

ffjc i e n t

– S

e e m s t

• i

d e n t i f y b e t t e r t h e v e r t i c a l s t a c k s

SLIDE 12

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 1 2

E x p e r i me n t a l r e s u l t s

• P

r e c i s i

• n

, R e c a l l , A c c u r a c y

• P
• r

p r e c i s i

• n

f

• r

P e a r s

• n

c

• r

r e l a t i

• n

Pearson Spearman

SLIDE 13

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 1 3

C

• mp

a r i s

• n
• S

p e a r ma n i s a c l e a r w i n n e r

– H

i g h e r a c c u r a c y

– B

e t t e r s t a b i l i t y w . r . t . T h r e s h

• l

d

SLIDE 14

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 1 4

S e n s i t i v i t y t

• s

a mp l i n g p e r i

• d
• S

mo

• t

h i n g e fg e c t

• f

s a mp l i n g f r e q u e n c y

– R

e d u c e d m a x i m u m a c c u r a c y

– I

n c r e a s e d s t a b i l i t y w . r . t . T h r e s h

• l

d

SLIDE 15

I n f Q

• O

c t , 2 5 2 1 6

• T

a

• r

m i n a 1 5

C

• n

c l u s i

• n

s

• E

n e r g y ma n a g e me n t i n c l

• u

d d a t a c e n t e r s

– N

e e d t

• c
• n

s i d e r n e t w

• r

k i n t e r a c t i

• n

s

– N

• p

e r

• d

e s t i n a t i

• n

/ p e r s

• u

r c e b r e a k d

• w

n

• f

t r a ffj c

• P

r

• p
• s

a l

• f

a n

• v

e l me t h

• d
• l
• g

y

– I

n t e r a c t i n g V M s f r

• m

a g g r e g a t e d n e t w

• r

k d a t a

– H

• r

i z

• n

t a l r e p l i c a t i

• n

+ t r a c e s n

• t

s y n c h r

• n

i z e d

• E

x p e r i me n t s

• n

a c l

• u

d i n f r a s t r u c t u r e

– C

• m

p a r i s

• n
• f

c

• r

r e l a t i

• n

i n d e x e s

– S

e n s i t i v i t y t

• s

a m p l i n g f r e q u e n c y

SLIDE 16

A Correlation-based Methodology to Infer Communication patterns between Cloud Virtual Machines Claudia Canali Riccardo Lancellotti

Dept of Engineering “Enzo Ferrari” University of Modena and Reggio Emilia