A Framework for Satisfying the Performance Requir irements of - - PowerPoint PPT Presentation

▶

Mar 25, 2023 167 likes •471 views

A Framework for Satisfying the Performance Requir irements of Contain inerized Software Systems Through Mult lti-Versionin ing Alireza Goli Cor-Paul Bezemer Hamzeh Khazaei Sara Gholami 1 / 28 An example problem: The Sla lashdot Effect 2

SLIDE 1

1 / 28

A Framework for Satisfying the Performance Requir irements of Contain inerized Software Systems Through Mult lti-Versionin ing

Sara Gholami Alireza Goli Cor-Paul Bezemer Hamzeh Khazaei

SLIDE 2

An example problem: The Sla lashdot Effect

2 / 28

SLIDE 3

One solu lution to the Sla lashdot Effect is is to in increase the resources

3 / 28

SLIDE 4

Another solu lution is is to manage the availa ilable le resources better

4 / 28

SLIDE 5

Hig igh-level architecture of f a serv ervice in in Docker, where req equests are lo load balanced in in a Round Robin manner

5 / 28

SLIDE 6

We present Docker wit ith mult lti-versioning: DockerMV

6 / 28

SLIDE 7

DockerMV takes away the service management nig ightmare

7 / 28

Load Balancer Load Balancer Load Balancer Load Balancer

SLIDE 8

8 / 28

How do we evaluate DockerMV?

SLIDE 9

The TeaStore applic lication, an onli line webstore applic lication

9 / 28

SLIDE 10

The Znn applic lication, a three-tie ier onlin line news appli lication

10 / 28

SLIDE 11

We conducted three exp xperim iments for the TeaStore appli lication

11 / 28

Ideal case experiment Recommender with multiple training (Only heavy weight) Adaptive experiment Adaptive load distribution (Mix of heavy and light weight) Worst case experiment Recommender with single training (Only light weight)

SLIDE 12

We conducted three exp xperim iments for the Znn applic lication

12 / 28

Ideal case experiment Multimedia responses only (Only heavy weight) Adaptive experiment Adaptive load distribution (Mix of heavy and light weight) Worst case experiment Text responses only (Only light weight)

SLIDE 13

Workload appli lied to the TeaStore appli lication

13 / 28

100 users sending HTTP requests for 1,000 seconds (Almost 97 requests per second) We defined the SLA response time threshold to be 450 ms

SLIDE 14

In the TeaStore application’s ideal case experiment, the response time exceeds th the th threshold under r th the lo load

14 / 28

Ideal case exp.

SLIDE 15

In In th the TeaStore application's worst t case experiment, , th the res esponse tim time fall lls below th the th threshold under th the loa load

15 / 28

Worst case exp.

SLIDE 16

In the TeaStore application’s adaptive experiment, the response time is main intained clo lose to th the th threshold

16 / 28

Adaptive exp.

SLIDE 17

The ratio io of f req equests res esponded by th the Heavy vyWeight version of f th the Recommender se serv rvic ice in in th the TeaStore's adaptive experiment

17 / 28

SLIDE 18

Workload appli lied to the Znn appli lication

18 / 28

SLIDE 19

In the Znn application’s ideal case experiment, the response time exceeds th the th threshold under r th the lo load

19 / 28

Ideal case exp.

SLIDE 20

In In th the Znn application's worst case experiment, th the res esponse tim time fall lls below th the th threshold under th the loa load

20 / 28

Worst case exp.

SLIDE 21

In the Znn application’s adaptive experiment, the response time is main intained around th the th threshold

21 / 28

Adaptive exp.

SLIDE 22

DockerMV source code is is public licly availa ilable le on Git itHub

22 / 28

https://github.com/pacslab/DockerMV

SLIDE 23

We present Docker wit ith mult lti-versioning: DockerMV

23 / 28

SLIDE 24

22 / 28

SLIDE 25

In the TeaStore application’s adaptive experiment, the response time is clo lose to th the th threshold

25 / 28

Adaptive exp.

SLIDE 26

26 / 28

SLIDE 27

In the Znn application’s adaptive experiment, the response time is main intained around th the th threshold

27 / 28

Adaptive exp.

SLIDE 28

28 / 28

Sara Gholami sgholami @ualberta.ca

SLIDE 29

Format of rule les for the lo load bala lancer

29 / 28

$METRIC $OPERATOR $THRESHOLD, (version $VERSION_NAME perc =$PERCENTAGE;)+ For example: RT > 0.4, version recommender:HeavyWeight perc=40; version recommender:LightWeight perc=60;

SLIDE 30

How to run DockerMV

30 / 28

docker service create [$OPTIONS] $IMAGE1 $REPLICATION1 … $IMAGEn $REPLICATIONn For example, docker service create e REGISTRY_HOST=host_ip e REGISTRY_PORT=1000 10.2.5.26 Network recommender 8080 rules.txt sgholami/teastore-recommender:HeavyWeight 1 sgholami/teastore-recommender:LightWeight 1