Top Banner
Benson Wu, 2005 1 Research Roadmap on network security: from practical firewall to anti-spam/spy ware PhD Candidate: Ming-Wei (Benson) Wu, 吳吳吳 Dept. of Electrical Engineering National Taiwan University [email protected] http:// www.ee.ntu.edu.tw/~benson
20

Research Roadmap on network security : from practical firewall to anti-spam / spyware

Jan 01, 2016

Download

Documents

raya-walls

Research Roadmap on network security : from practical firewall to anti-spam / spyware. PhD Candidate: Ming-Wei (Benson) Wu, 吳明蔚 Dept. of Electrical Engineering National Taiwan University [email protected] http://www.ee.ntu.edu.tw/~benson. Questions to Answer. Changes in Security - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 1

Research Roadmap on network security:

from practical firewall to anti-spam/spyware

PhD Candidate: Ming-Wei (Benson) Wu, 吳明蔚Dept. of Electrical Engineering

National Taiwan University

[email protected]

http://www.ee.ntu.edu.tw/~benson

Page 2: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 2

Questions to Answer

Changes in Security Perimeter Depth Granularity

Case studies Anti-spyware

Conclusions

What have I done? Brief background Research

What does Internet Security look like today?

Changes in Internet users and applications

Changes in Threat Legacy security measures

Page 3: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 3

Leadership Implementation

Brief Background

工研院交大網路測試中心Network Benchmarking Lab

交大資科高速網路實驗室High Speed Network Lab

利基網路L7 Networks

資訊工業策進會Information Industry

Institution

台大電機分散式網路實驗室Dependable and

Distributed Network Lab

台網資訊中心台灣新世代網路菁英

TaiWan Internet Next Generation

2000~20032000~2003

2003~20052003~2005Domain knowledge

Discipline

馬尼拉美國學校ISM

1992~19961992~1996English

Page 4: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 4

Research

Connectivity

P2P GatewayWeb ServicesDigital Home

Security

Security GatewayBenchmarking XML FirewallAnti-spyware

Anti-spam

Public interests…

Open Source Dev.Textbook writingMag. article writing

Page 5: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 5

Internet Evolution

Changes in Networking Technologies Changes in Internet Users and Internet

Applications Changes in Security Accessories

Page 6: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 6

Changes in Internet Applications: Primitive Web becoming Web Services

Is Primitive Web enough? When they are still newbie…they want to “join”

ALL Client-to-Server When they become big enough…they want to

“share” Some Peer-to-Peer (P2P) Some Server-to-Server (Web Services)

Page 7: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 7

Changes in Internet Users:from Browsing towards Clicking

Necessary services at one-click: Web Services

e.g. One-stop shopping

Necessary authentications at one-time: Single Sign-On

e.g. One-click cart/basket

Necessary confidentiality with higher-granularity: XML Enc.

e.g. Interleaved workflow

Page 8: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 8

Such changes are more like a reality…

Some numbers about P2P 2 millions of Kuro users and 50.2% of teenag

ers (15~22) have visited either Kuro or EZPeer (創市際市場研究顧問公司 , 2003/09)

Some NT$9.6 billions lost due to P2P sharing (資策會網路通訊雜誌 , 2003/06)

Some numbers about Web Services… 79% are evaluating (Accenture) 52% are using or testing (TechMetrix) 45.5% consider security to be the biggest obs

tacle (BusinessWeek)

Page 9: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 9

The Evolution of P2P: Darwinism

Page 10: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 10

2004 P2P Popularity and User Rating

Top 20 Popular P2P File-Sharing Applications

0

50,000,000

100,000,000

150,000,000

200,000,000

250,000,000

300,000,000

350,000,000

400,000,000

# of D

ownlo

ads

0102030405060708090100

TotalDownloads

User Rating

Page 11: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 11

Extending Client-Server to P2P:Its Problems and Solutions

Connectivity Internet transparency?

How to connect resources successfully? Sol: middleman (e.g. gatekeeper in H.323, broker in middleware,

renderzvous node in JXTA)

Scalability size?

How to locate MANY resources? Sol: smart routing (make use of DHT)

time? How to locate resources INSTANTLY?

Sol: Distributed hash table or DHT (resilience?)

Page 12: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 12

Extending Client-Server to Web Services:Its Problems and Solutions

What most XML firewall do? How to manipulate only parts of a document?

Per-element XML encryption/signing How to authenticate/authorize between more than

two parties Single-Sign On

How to assure the validity of Web Services’ action? SOAP Schema validation SOAP Digital Signature verification

Page 13: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 13

Changes in Threat:Volume and Impact Security is tougher than ever

In volume: >600% 137,529 reported incidents during 2003, which is more

than 6 times of 2000 (CERT) In impact: <10 minutes

SQL slammer (aka. Sapphire) happened to own the Internet in less than 10 minutes in 2003

Page 14: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 14

A Remind of Legacy Security Measures

Access security Firewall Content Filter

Data security Virtual Private Network (VPN)

System security Intrusion Detection System (IDS) Antivirus

Page 15: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 15

Technical Analysis: Issues

FW: must leave alone well-known ports, e.g. 80 IDS: false alarm, new attack, correlation AV: new virus, signatures, where (desktop or

network), polymorphism CF: false positives, false negatives VPN: management overhead, interoperability

Page 16: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 16

Changes in Security: Perimeter, Depth and Granularity

Existing security measures that protect you TCP/IP firewall: packet-level Virtual Private Network (VPN): IP-level tunneling Content filter: application-level Intrusion Detection System (IDS): application-level Antivirus: application-level

Situation had changed Network perimeters have become less defined due to pervasive mobile devices

(e.g. WLAN, PDA, etc.) 80% of all attacks come from external parties, yet 80% of all security-related loss

es are due to remaining 20% of attacks Increasing Depth

Stand-alone security measure Integrated all-in-one approach Demand for internal security is emerging (plus more applications and more u

sers requiring higher bandwidth) Finer Granularity

Packet-level Application-level Per-flow basis Per-element basis

Page 17: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 17

Anti-spyware: What are we dealing?

Spyware Definition: a generic term referring to a class of software program

s that could violate and potentially jeopardize people privacy and security concerns

Examples: Gator, Cydoor, Aureate, Comet Cursor and Web3000 could be found in many free applications (Kazaa, Bearshare, iMesh and Limewire) Read the EULA (End-user license agreement)

How serious? nearly 70% spyware penetration in campus environment (Saroiu et al., 2004)

Impact: credit card numbers could be stolen keystrokes could be captured browser settings could be modified users could be profiled …following spyware often comes with Trojan, virus and worms

Page 18: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 18

Anti-spyware: Rootkits as an example

Definition: software that comprise tools to erase traces of the intrusion from audit logs have "backdoors" that allow easy access hide the rootkit itself from administrators

Types: User-mode rootkit

replacing system binaries with trojaned ones Kernel-mode rootkit (with Linux Kernel Module support)

insert a module that overrides kernel syscalls Runtime kernel patchings

writing to /dev/kmem (with or without the LKM support) Tools for Rootkit Detection

Tripwire AIDE (Advanced Intrusion Detection Environment) Chkrootkit (~56 rootkits)

Page 19: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 19

Conclusions

Firewall Application-aware filtering Anti-spam Single-sign on

IDS IPS

VPN SSL VPN

Anti-virus Anti-spyware

Page 20: Research Roadmap on  network security :  from  practical firewall  to  anti-spam / spyware

Benson Wu, 2005 20

Many thanks for your time :)