• No results found

Cloud Analytics for Capacity Planning and Instant VM Provisioning

N/A
N/A
Protected

Academic year: 2021

Share "Cloud Analytics for Capacity Planning and Instant VM Provisioning"

Copied!
27
0
0

Loading.... (view fulltext now)

Full text

(1)

Cloud Analytics for Capacity Planning

and Instant VM Provisioning

Yexi Jiang

Florida International University

Advisor: Dr. Tao Li

(2)

Presentation Outline

Background

Cloud Capacity Prediction

Predict provisioning resource demand

Estimate de-provisioning requests

Experimental evaluation results

Instant Cloud Provisioning

Predict VM provisioning demand

Experimental evaluation results

(3)

Background

• What is

Cloud Analytics

?

Rapidly identify cloud resource or application trouble spots so you can

solve the problem.

What is the objective of cloud analytics?

The cloud platform itself.

• What can cloud analytics do?

– Workload analysis

– System fault diagnostics

(4)

Smart Cloud Enterprise trace data

5 month, 35k+ requests, 120+ image types, 20+ features each record

Important Features: Image Name, Owner, Start Time, End Time, ID

(5)

Aggregating the Raw Data

weekly

daily

hourly

Cannot reflect

real capacity

Just right

(6)

Aggregating the Raw Data

Measurement Weekly Daily Hourly

Coefficient of Variance (CV) 0.5606 0.7915 1.2249 Skewness 0.3295 1.5644 5.4464 Kurtosis 1.62 5.8848 52.4103

weekly

daily

hourly

Cannot reflect

real capacity

Just right

Too irregular

(7)

Presentation Outline

Background

Cloud Capacity Prediction

Predict provisioning resource demand

Estimate de-provisioning requests

Experimental evaluation results

Instant Cloud Provisioning

Predict VM provisioning demand

Experimental evaluation results

(8)

Cost of Data Centers

31% of the cost is related to power.

As hardware price continuously decreases, the proportion would

further increase.

The US EPA estimates the energy usage at data centers is experiencing

successive doubling every five years. (7.4 billion in 2011)

(9)

Motivation

• Reduce power cost via capacity prediction

Cos

t of the

Cl

oud

P

ro

vi

de

r

Prepared Resource

Real Requirement

(10)

Motivation

• Reduce power cost via capacity prediction

Cos

t of the

Cl

oud

P

ro

vi

de

r

Prepared Resource

Predicted Resource

Real Requirement

(11)

Candidate Time Series

Capacity time series

Non-stationary.

Difficult to model directly

Provisioning /de-provisioning

time series

Obvious temporal pattern

(12)

Basic Idea

• Capacity = (# existing VMs) + (# provisioning) - (# de-provisioning)

Existing VM in

cloud

-+

Predicted

Provisioning

Predicted

De-provisioning

Predicted Capacity

(13)

Predicting Provisioning

Demands

• Ensemble method for time series prediction

Individual prediction techniques used:

Moving Average. Naïve predictor.

Auto Regression. Linear predictor. – Neural Network. Non-linear predictor.

Gene Expression Programming. Genetic algorithm.

Support Vector Machine. Linear predictor with non-linear kernel.

• Dynamic weighted linear combination

• Weight update

w

p(t)

weight of predictor p

v

p

predicted value of individual

predictor p

c

p(t)

cost of predictor p at time t

(14)

Cloud Prediction Cost

Over-prediction: cost of resource waste.

R

function:

Under-prediction: cost of SLA penalty.

T

function:

Property: Non-negative, Monotonic.

))

(

~

),

(

(

))

(

~

),

(

(

v

t

v

t

T

v

t

v

t

R

C

=

+

(15)

Prediction Result

• Ensemble has the best average performance.

(16)

Predicting De-provisioning

• Use the life span CDF

F(x)

of VMs to estimate number of

de-provisioning requests

• Estimation of distribution: step-wise function.

(17)

De-provisioning evaluation

Test data:

last 60 day.

Test methods:

1. No preparation at all (None)

2. Always prepare the maximum capacity

(Maximum)

3. Time series prediction (Time Series)

4. Life span distribution despite of image

60 days of data (Dist 60)

90 days of data (Dist 90)

• Global distribution estimation method outperforms the time

series prediction method.

(18)

Presentation Outline

Background

Cloud Capacity Prediction

Predict provisioning resource demand

Estimate de-provisioning requests

Experimental evaluation results

Instant Cloud Provisioning

Predict VM provisioning demand

Experimental evaluation results

(19)

Motivation

Problem

:

Existing clouds are not “instant”, not suitable for mid-job

scaling and urgent tasks.

VM preparation is fast, but patching, security assurance, manual process and

other processes cost time.

Known solutions

:

Prepare extreme large number of different types of VMs.

Waste

resource

Ask customers to provide schedule.

Impractical

Our Idea

: Make good use of the customer historical requests to infer

(20)

Core Idea

Model and

predict

demands

Predict

Results

Pre-provision

at suitable

time

Wait for

Requests

Assign VMs

to

customers

(21)

Focus on individual types

• No obvious temporal patterns for individual image type.

Ensemble is still

required.

(22)

Focus on popular VM types

1) About 10% (12) of the 124 VM types consists more than 80% requests

2) Inflection point divides the VM types into popular group and rare group

3) Requests for rare image types appear randomly.

(23)
(24)

Experimental Evaluation

• Ensemble method have the best performance in

reducing waiting time and resource waste.

(25)

Conclusion

Capacity Prediction

The demand of cloud capacity can be estimated by predicting provisioning and

de-provisioning requests

Use time series ensemble method for provisioning prediction

Use VM life span model for de-provisioning prediction

Instant cloud provisioning

Pre-provision VMs before requests arrive

Predict VM provision requests use time series ensemble method

The average provisioning fulfillment time can be reduced by 85%+

Future work

Improve prediction with user profile

(26)
(27)

Thank you

Related Paper:

Intelligent Cloud Capacity Management. (NOMS 2012)

ASAP: A Self-Adaptive Prediction System for Instant Cloud

Resource Demand Provisioning. (ICDM 2011)

Patent:

References

Related documents

One approach to points-to analysis is quite similar to control flow analysis. For each variable named id we introduce a set variable [[ id ]] ranging over the possible pointer

A survey of risk management strategies practised by NHS hospitals when transferring critically ill patients, found that only 29.3% of hospitals reported using a dedicated

In this section we formulate an algebraic proof system OFPC that operates with noncommutative polynomials in which every monomial is a product of variables in nondecreasing order

Table 3 Computational burden associated with the RML, SGN, AML and AGB algorithms for a single iteration of a second order adaptive notch

The Capability Integra- tion and Experimentation Coordination Cell (CIECC) will provide this control in exercise TRIDENT JUNCTURE 15 (TRJE 15/Part 1) as a mission tailored

CHANGES AND CONFUSION BROUGHT ABOUT BY THE DYKES DECISION Ever since Dykes, 8 Indiana's appellate courts have been con- fused over what constitutes an accident in

Tongan politicians conducted part of their campaign in New Zealand because even though Tongans living in New Zealand cannot vote in Tonga, they used new media to influence