effects of locality, content and java runtime on video performance vikram chhabra, akshay kothare,...

23
EFFECTS OF LOCALITY, EFFECTS OF LOCALITY, CONTENT AND JAVA RUNTIME CONTENT AND JAVA RUNTIME ON VIDEO PERFORMANCE ON VIDEO PERFORMANCE Vikram Chhabra, Akshay Kothare, Vikram Chhabra, Akshay Kothare, Mark Claypool Mark Claypool Computer Science Department Computer Science Department Worcester Polytechnic Institute Worcester Polytechnic Institute Worcester, Massachusetts, USA Worcester, Massachusetts, USA

Upload: bruce-marvin-stevens

Post on 31-Dec-2015

218 views

Category:

Documents


1 download

TRANSCRIPT

EFFECTS OF LOCALITY, EFFECTS OF LOCALITY, CONTENT AND JAVA RUNTIMECONTENT AND JAVA RUNTIME

ON VIDEO PERFORMANCEON VIDEO PERFORMANCE

Vikram Chhabra, Akshay Kothare, Vikram Chhabra, Akshay Kothare,

Mark ClaypoolMark ClaypoolComputer Science DepartmentComputer Science Department

Worcester Polytechnic InstituteWorcester Polytechnic Institute

Worcester, Massachusetts, USAWorcester, Massachusetts, USA

Introduction

• Increased bandwidth has enabled video streaming to the desktop.

• Java’s “Write once, run anywhere” - preferred choice for multimedia applications.

Java Runtime

Java Source

Bytecode

JVM

Native OS

JIT Compiler

C++ Source

Java Performance

• Java lags the speed of C / C++.

• Need to explore factors influencing the performance of Java.

Performance of Multimedia

• Frame Rate : Numbers of frames displayed per second.

• Jitter : Variance in time taken for a frame to travel from server to client.

Previous Work

• Most performance evaluations have been for traditional media.

• Our previous work suggests:

– Local access of media

– Processing power of the native hardware

mostly influence the performance of Java.

Our Approach

• Setup a controlled environment

• Serve movie to Java client

• Observe frame arrival time

• Calculate frame rate & jitter

• Derive conclusions

Experimental Setup

C++ Server Java ClientTCP/IPNo frame loss

Hard Disk

Parameters to be tested

• Network distance

• Movie content

• JIT compilation

Parameter One : Network Distance

Four Setups:

1. Local Playback

2. Dedicated Client/Server

3. Normal Setup

4. Different SwitchFixed Parameters :

Operating System : Linux 6.3 kernel version 2.2.13.

Server : Pentium MMX 233 MHz PC with 64MB of RAM.

Client : Pentium II 300 MHz PC with 128MB of RAM.

Network : Ethernet broadband coaxial cable at 10Mbps.

Movie : A space scene, encoded as 120 “ I ” frames

Parameter Two : Movie Content• Ten “ perceptually different ” movies

– fixed resolution of 200 x 200 pixels

– 120 frames each, 4 seconds each

– “I” and “IBBPBBPBBPBB” group of pictures format.

• I Frame : Fully Encoded

• B or P Frames : Relatively Encoded

• Fixed Parameters :Client : WinNT 4.0 service pack 6

(Pentium MMX 233 MHz, 64MB RAM)

Server : SuSE Linux 6.4 Kernel version 2.2.14

(Intel PIII 500Mhz PC, 128 MB RAM)

Network : 100 Mbps Ethernet

The Movies

Parameter Three : JIT Compilation

• JIT “on” vs. “off”

• “Java” vs. “C”

Fixed Parameters :

Operating System : Linux 6.3 kernel version 2.2.13.

Client : Pentium II 300 MHz PC with 128MB of RAM.

Movie : A space scene, encoded as 120 “ I ” frames

RESULTS

Network Distance – Frame Rate

0.00

0.200.40

0.600.80

1.001.20

1.401.60

1.802.00

Local Dedicated Normal Switch

Distance between Client and Server

Fram

e R

ate

(fps)

.

“ Local media access faster than different network setups ”

Network Distance – Jitter

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

Local Dedicated Normal Switch

Distance between Client and Server

Jitt

er (

mea

sure

d as

the

sta

ndar

d

dev

iatio

n in

del

ay)

“ Local media access shows lesser jitter than network access ”

Movie Content – Frame Rate

0.00

1.00

2.00

3.00

4.00

5.00

6.00F

ram

es p

er s

eco

nd I Frames

IBB(PBB)*4

“ Movie content affects the frame rate ”

Movie Content – Jitter

020406080

100120140160

Jitt

er (

Mea

sure

d as

the

Sta

nda

rd

Dev

iati

on in

del

ay)

I Frames IBB(PBB)*4

“ Movie content affects jitter ”

JIT Compilation – Frame Rate

0

5

10

15

20

25

No JIT JIT C

Fra

mes

per

sec

ond

“ JIT enhances performance, Java still lags power of C ”

JIT Compilation – Jitter

0

10

20

30

40

50

60

70

80

No JIT JITJitt

er (

mea

sure

d as

the

Sta

ndar

d D

evia

tion

in d

elay

)

“ JIT reduces jitter ”

Conclusions

• Local media access faster than network– 5% higher frame rate, 25% lesser jitter

• Not much difference over different networks

• Movie content has significant implications– Blue movie 3 times faster and shows 40%

lesser jitter than Mixbag

Conclusions (contd.)

• JIT gives performance boost – 100 % increase in frame rate – 25% decrease in jitter

• Java still lags C– Java (JIT on) frame rate 10 times less than C

Future Work

• Trying different protocols – UDP, RMI, CORBA instead of TCP.

• Finding relationship between decrease in frame rate with perceptually different movie content.

• Trying different JVMs and/or JITs.