Fuzzy Control meets Software Engineering

Fuzzy Control meets Software Engineering

Pooyan Jamshidi

IC4- Irish Centre for Cloud Computing and Commerce

School of Computing, Dublin City University

[email protected]

Dagstuhl SeminarControl Theory meets Software EngineeringSeptember, 2014

Naeem Esfahani and Sam Malek, “Uncertainty in Self-Adaptive Software Systems”

Knowledge SpecificationUncertainty

2

3

Application #1:

Auto-scaling

~50% = wasted hardware

Actual traffic

Typical weekly traffic to Web-based applications (e.g., Amazon.com)

5

Problem 1: ~75% wasted capacityActual

demand

Problem 2:

customer lost

Traffic in an unexpected burst in requests (e.g. end of year traffic to Amazon.com)

6

Really like this??

Auto-scaling enables you to realize this ideal on-demand provisioning

Time

Dem

and

?Enacting change in the Cloud resources are notreal-time

7

Capacity we can provision with Auto-Scaling

A realistic figure of dynamic provisioning

8

9

0 50 1000

500

1000

1500

0 50 100100

200

300

400

500

0 50 1000

1000

2000

0 50 1000

200

400

600

10

11

These quantitative values are required to be determined by the user requires deep

knowledge of application (CPU, memory, thresholds)

requires performance modeling expertise (when and how to scale)

A unified opinion of user(s) is required

Amazon auto scalingMicrosoft Azure Watch

12

Microsoft Azure Auto-

scaling Application Block

12

13

Uncertainty related to enactment latency: The same scaling action (adding/removing a VM with precisely the same size) took different time to be enacted on the cloud platform (here is Microsoft Azure) at different points and this difference were significant (up to couple of minutes).The enactment latency would be also different on different cloud platforms.

14

Offline benchmarking

Trial-and-error

Expert knowledge

Costly and

not systematicA. Gandhi, P. Dube, A. Karve, A. Kochut, L. Zhang, Adaptive, “Model-driven Autoscaling for Cloud Applications”, ICAC’14

arrival rate (req/s)

95

% R

esp

. tim

e (m

s)

400 ms

60 req/s 15

RobusT2Scale Initial setting + elasticity rules + response-time SLA

environment monitoring

application monitoring

scaling actions

Fuzzy ReasoningUsers

Prediction/Smoothing

16

0 0.5 1 1.5 2 2.5 30

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

Region of

definite

satisfaction

Region of

definite

dissatisfaction Region of

uncertain

satisfaction

Performance Index

Pos

sibi

lity

Performance IndexPos

sibi

lity

words can mean different things to different people

Different users often recommend different elasticity policies

0 0.5 1 1.5 2 2.5 30

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

Type-2 MF

Type-1 MF

17

18

Workload

Response time

0 10 20 30 40 50 60 70 80 90 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

x2

uM

embe

rship

gra

de

0 10 20 30 40 50 60 70 80 90 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

x2

uM

embe

rship

gra

de

=>

=>

mean

sd

19

Rule (𝒍)

Antecedents Consequent

𝒄𝒂𝒗𝒈𝒍

Workload Response-

time Normal

(-2) Effort

(-1)

Medium Effort

(0)

High Effort (+1)

Maximum Effort (+2)

1 Very low Instantaneous 7 2 1 0 0 -1.6 2 Very low Fast 5 4 1 0 0 -1.4 3 Very low Medium 0 2 6 2 0 0 4 Very low Slow 0 0 4 6 0 0.6 5 Very low Very slow 0 0 0 6 4 1.4 6 Low Instantaneous 5 3 2 0 0 -1.3 7 Low Fast 2 7 1 0 0 -1.1 8 Low Medium 0 1 5 3 1 0.4 9 Low Slow 0 0 1 8 1 1

10 Low Very slow 0 0 0 4 6 1.6 11 Medium Instantaneous 6 4 0 0 0 -1.6 12 Medium Fast 2 5 3 0 0 -0.9 13 Medium Medium 0 0 5 4 1 0.6 14 Medium Slow 0 0 1 7 2 1.1 15 Medium Very slow 0 0 1 3 6 1.5 16 High Instantaneous 8 2 0 0 0 -1.8 17 High Fast 4 6 0 0 0 -1.4 18 High Medium 0 1 5 3 1 0.4 19 High Slow 0 0 1 7 2 1.1 20 High Very slow 0 0 0 6 4 1.4 21 Very high Instantaneous 9 1 0 0 0 -1.9 22 Very high Fast 3 6 1 0 0 -1.2 23 Very high Medium 0 1 4 4 1 0.5 24 Very high Slow 0 0 1 8 1 1 25 Very high Very slow 0 0 0 4 6 1.6

Rule

(𝐥)

Antecedents Consequent

𝒄𝒂𝒗𝒈𝒍Work

load

Response

-time-2 -1 0 +1 +2

12 Medium Fast 2 5 3 0 0 -0.9

10 experts’ responses

𝑅𝑙: IF (the workload (𝑥1) is 𝐹𝑖1, AND the response-time (𝑥2) is 𝐺𝑖2), THEN (add/remove 𝑐𝑎𝑣𝑔

𝑙 instances).

𝑐𝑎𝑣𝑔𝑙 =

𝑢=1𝑁𝑙 𝑤𝑢

𝑙 × 𝐶

𝑢=1𝑁𝑙 𝑤𝑢

𝑙

Goal: pre-computations of costly calculations to make a runtime efficient elasticity reasoning based on fuzzy inference 20

Liang, Q., Mendel, J. M. (2000). Interval type-2 fuzzy logic systems: theory and design. Fuzzy Systems, IEEE Transactions on, 8(5), 535-550.

Scaling Actions

Monitoring Data

21

0 10 20 30 40 50 60 70 80 90 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.5954

0.3797

𝑀

0 10 20 30 40 50 60 70 80 90 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.2212

0.0000

0 10 20 30 40 50 60 70 80 90 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

x2

u

0 10 20 30 40 50 60 70 80 90 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

x2

uMonitoring data

Workload

Response time22

0 10 20 30 40 50 60 70 80 90 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.5954

0.3797

0 10 20 30 40 50 60 70 80 90 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0.95680.9377

23

24

𝑦𝑙 , 𝑦𝑟

25

26

0 50 1000

500

1000

1500

0 50 100100

200

300

400

500

0 50 1000

1000

2000

0 50 1000

200

400

600

0 50 1000

500

1000

0 50 1000

500

1000

27

28

0 10 20 30 40 50 60 70 80 90 100-500

0

500

1000

1500

2000

Time (seconds)

Num

ber

of

hits

Original data

betta=0.10, gamma=0.94, rmse=308.1565, rrse=0.79703



0

0.2

0.4

0.6

0.8

1

1.2

1.4

Big spike Dual phase Large variations Quickly varying Slowly varying Steep tri phase

0 50 1000

500

1000

1500

0 50 100100

200

300

400

500

0 50 1000

1000

2000

0 50 1000

200

400

600

0 50 1000

500

1000

0 50 1000

500

1000

Roo

t Rel

ativ

e Squ

are

d E

rror

29

SUT Criteria Big spike Dual phaseLarge

variations

Quickly

varying

Slowly

varying

Steep tri

phase

RobusT2Scale𝑟𝑡95% 973ms 537ms 509ms 451ms 423ms 498ms

𝑣𝑚 3.2 3.8 5.1 5.3 3.7 3.9

Overprovisioning𝑟𝑡95% 354ms 411ms 395ms 446ms 371ms 491ms

𝑣𝑚 6 6 6 6 6 6

Under

provisioning

𝑟𝑡95% 1465ms 1832ms 1789ms 1594ms 1898ms 2194ms

𝑣𝑚 2 2 2 2 2 2

SLA: 𝒓𝒕𝟗𝟓 ≤ 𝟔𝟎𝟎𝒎𝒔For every 10s control interval

•RobusT2Scale is superior to under-provisioning in terms of guaranteeing the SLA and does not require excessive resources•RobusT2Scale is superior to over-provisioning in terms of guaranteeing required resources while guaranteeing the SLA 30

0

0.02

0.04

0.06

0.08

0.1

alpha=0.1 alpha=0.5 alpha=0.9 alpha=1.0

Roo

t M

ean S

quare

Err

or

Noise level: 10% 31

Application #2: Self-Adaptive

Software Connectors

⊭ 𝑅

Environment=D

Environment=D’

Environment=D’

⊨ 𝑅

⊨ 𝑅

Adapted to satisfy requirements while it is running

Reliable Runtime Efficient

33

00.5

11.5

22.5

33.5

44.5

5

00.5

11.5

22.5

33.5

44.5

5

1.5

2

2.5

3

3.5

4

input2input1

outp

ut1

34

35

0

1

2

3

4

5

x 10-3

1 2 3 4 5 6 7 8

0

0.05

0.1

0.15

0.2

0.25

1 2 3 4 5

36

0.05

0.1

0.15

0.2

0.25

0.3

0.35

Type-1 FLS Type-2 FLS

RM

SE

• The rule reduction reduced the rules quite considerably.

• IT2 FLCs are more robust due to less mean error and less variation in the estimation error.

• T1 FLCs in some realization drop more rules in comparison with the IT2 FLCs.

• IT2 FLC original designs can be designed with less rules.

37

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

1% 2% 3% 4% 5% 6% 7% 8% 9% 10%

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1% 2% 3% 4% 5% 6% 7% 8% 9% 10%

0

0.2

0.4

0.6

0.8

1

1% 2% 3% 4% 5% 6% 7% 8% 9% 10%

Noise Amplitude (α=0.5)

RM

SE

α=0.95α=0.7

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1% 2% 3% 4% 5% 6% 7% 8% 9% 10%

T1 FLCIT2 FLC

38

Future Directions

Current Solution(my PhD Thesis+ IC4 work on auto-scaling)

Future PlanUpdating Kin MAPE-K@ RuntimeDesign-time

Assistance

Multi-cloud

41

42

Challenge 1: ~75% wasted capacityActual

demand

Challenge 2:

customer lost

⊭ 𝑅

Environment=D

Environment=D’

Environment=D’

⊨ 𝑅

⊨ 𝑅

Adapted to satisfy

requirements

while it is running

Reliable

Run-time Efficient

Current Solution(my PhD Thesis+ IC4 work on auto-scaling)

Future PlanUpdating Kin MAPE-K@ RuntimeDesign-time

Assistance

Multi-cloud

RobusT2Scale Initial setting + elasticity rules + response-time SLA

environment monitoring

application monitoring

scaling actions

Fuzzy ReasoningUsers

Prediction/Smoothing

43

http://computing.dcu.ie/~pjamshidi/PDF/SEAMS2014.pdf

More Details?

=> http://www.slideshare.net/pooyanjamshidi/

Slides?

=>

44

Thank you!

=>

Aakash Ahmad Claus Pahl Soodeh Farokhi

Requirement: R

0 50 1000

500

1000

1500

0 50 100100

200

300

400

500

0 50 1000

1000

2000

0 50 1000

200

400

600

0 50 1000

500

1000

0 50 1000

500

1000

Environment: D

Model: S

MonitoringImplementation

Execution

Runtime

Design-time

Reasoning

Self-adaptation

Specification

S,D⊨R

S,D⊭R

46

Looking for references?http://computing.dcu.ie/~pjamshidi/PDF/SEAMS2014.pdf

47

48

49

50

𝑦𝑟 − 𝑦𝑙

51

52