互联网反向工程
Reverse Engineer the Computer Networks
Hui Zhou, Yongji Wang
Institute of Software, Chinese Academy of Sciences Graduate University of Chinese Academy of Sciences
P.O.Box 8718, Beijing 100080, China
(8610)62661900 ext 8283
{hzhou, ywang }@itechs.iscas.ac.cn
ABSTRACT
Software reverse engineering has undergone many milestones and stepped from research to industry quickly in recent ten years. By analogy, we found that it is also possible to reverse engineering the computer networks. The goal of network reverse engineering is to annotate a map of the networks with properties such as node distribution, connectivity, and workload. It is very necessary, but also challenging, to reverse engineering the computer networks. We compared the reverse engineering of software and network domains from five fundamental perspectives: source, data analysis, presentation, validation, and prediction. Finally, two examples are evaluated to see how the reverse engineering of both software and network impact each other. The first example is RichMap, which applied the snapshot concept of software domain, is built to infer the topology and all link available-bandwidth of a network. The second example illustrates how the network reverse engineering can help accelerate the dissemination of software patch.
Categories and Subject Descriptors
C.2.1 [COMPUTER-COMMUNICATION NETWORKS]: Network Architecture and Design – Distributed networks, Network communications, Network topology.
General Terms
Experimentation, Measurement.
Keywords
Computer network, measurement, and reverse engineering.
1. INTRODUCTION
Software reverse engineering is, in practice, one of the most important endeavors in software engineering. This stems from the fact that software systems are complex and often poorly specified and documented. As a result, software practitioners need to spend a substantial amount of time understanding the source code from a structural and behavioral perspective, before carrying out any maintenance task. In this context, most reverse engineering processes follow the same pattern: A program is analyzed through static or dynamic analysis and the collected low-level program information is transformed into a higher level, more abstract presentation. The presentation helps engineers better understand the rationale of the code and thus facilitate future changes. From networking standpoint, reverse engineering is the process of analyzing a target network to identify the design of network and create presentations. Specifically, we take “design” to mean how its components, e.g. node, link and internal network, are assembled and configured, as well as how they operate.
It has become very necessary to reverse engineering the computer networks. Given the dynamic nature of the Internet, keeping track of network information manually is a daunting (if not impossible) task. Network operators generally can’t draw a complete map of their networks since many internal parts can undergo different scales of changes but will not report these changes immediately. Therefore, they need reverse engineering systems to detect underutilized or congested links, to plan network capacity upgrades, and to deploy security infrastructure. In addition, many users also need to verify whether they get the network service stated in their service-level agreements with the Internet service providers (ISPs). Bandwidth-sensitive applications, e.g. web video and peer-to-peer conference, also need to perceive the network internals so as to adjust their transmission strategies in a timely manner.
However, network reverse engineering is a challenging task. The key reason is that the design of the Internet can’t provide explicit support for end nodes to obtain information about the network internals. A network typically consists of many small networks; such networks are under different administrative control, so there is no single place from which one can obtain a complete picture of the specified target network. Furthermore, the Internet is so heterogeneous that an approach found to be useful in a certain networks may not be effective somewhere [1].
To reverse engineer the computer networks, we not only need to study network technology, but also need to learn from software engineering. We argue that a collaborative effort of software and network domains would achieve significantly more than the isolated efforts of individuals. In particular, we are required to address questions such as “can we reverse engineer the computer networks?”, “where are the reverse engineering of software and network different?”, “where are they the same?”, and “can they benefit from the collaborative effort?”
In the process of answering these questions, this paper makes two contributions. First, compared the reverse engineering in both software and networking fields, and found that they differ at four basic points: input, output, methodology, and validation. Second, we carried out experiments to illustrate that both software and network communities can benefit from the collaborative effort at reverse engineering.
The remainder of this paper is organized as follows. Section 2 summarized the related work on both software and network fields. Section 3 estimated the reverse engineering techniques of both software and networking, and then Section 4 presents our experiments on RichMap, which draws a series of streaming snapshots about designated networks. Finally, Section 5 discusses our findings, and Section 6 concludes the paper.
computer network Ⅰ .Introduction Computer networks,the wide spread sharing of information among groups of computers and their users,are a central part of ...
Computer Networks-Ch1-part1_IT/计算机_专业资料。计算机网络Chapter 1 Introduction What is a computer network? ? Computer Network: a collection of ...
Computer Networks, Chapter 1 4 Uses of Computer Networks ? ? Business Applications Client-Server Model Computer Networks, Chapter 1 5 Home Applications ? ...
What is Computer Networks ? Uses(应用) of Computer Networks ? Components of Computer Networks(Hardware and Software) ? Reference Model of Computer Networks...
Computer Networks and Communication Systems_专业资料。Abstract: Sensor Networks have been heralded as one of 21 most important technologies for 21 st ...
computer networks第四版习题答案_工学_高等教育_教育专区。computer networks 第...In the star configuration, the router is in the middle of floor 4. ...
andrew s. tanenbaum 《computer networks》fourth edition习题答案_工学_高等...17. the probability, pk , of a frame requiring exactly k transmissions is...
http://www.ceiva.com/ World’s smallest web server http://www-ccs.cs.umass.edu/~shri/iPic.html 13 Internet phones Category of Computer Networks ?...
1-2 What is computer networks? The term “computer network” means a collection of autonomous computers interconnected by a single technology. It is ...
暂无评价 117页 2财富值 Computer Networks(3) 暂无评价 86页 2财富值...What is the main circuit board of a computer? A. PC subsystem B. ...
热门文档