DMCA
Efficient Querying and Maintenance of Network Provenance at Internet-Scale (2010)
Cached
Download Links
Citations: | 48 - 26 self |
Citations
4465 | Chord: A scalable Peer-To-Peer lookup service for internet applications
- Stoica, Morris, et al.
- 2003
(Show Context)
Citation Context ... are a natural and compact way to implement a variety of routing protocols and overlay networks. For example, traditional routing protocols can be expressed in a few lines of code [17], and the Chord =-=[23]-=- distributed hash table in 47 lines of code [16]. When compiled and executed, these declarative networks perform efficiently relative to imperative implementations. The techniques proposed in this pap... |
1033 | Symbolic boolean manipulation with ordered binary decision diagrams
- Bryant
- 1992
(Show Context)
Citation Context ...he dimensions of granularity (domains defined according to nodes, tuples, or networks), distribution (centralized or distributed), and representation (using derivation trees, binary decision diagrams =-=[2]-=-, algebraic structures, etc.). We show that various distributed systems (in particular, diagnostics, network debugging, and distributed trust management) map naturally to network provenance. This pape... |
670 | Practical network support for IP traceback
- Savage, Wetherall, et al.
- 2000
(Show Context)
Citation Context ...rming network diagnostics, identifying malicious and misbehaving users, and enforcing trust management policies in distributed systems. Each goal has led to a series of application-specific proposals =-=[21, 1, 11, 24, 9, 12]-=- that focus on improving network support for accountability and providing efficient mechanisms to trace packets and information flows through the Internet. This paper explores the generic data managem... |
429 | Why and where: A characterization of data provenance
- Buneman, Khanna, et al.
- 2001
(Show Context)
Citation Context ...lis, Indiana, USA. Copyright 2010 ACM 978-1-4503-0032-2/10/06 ...$10.00. Keywords Provenance, Distributed query processing, Declarative networking 1. INTRODUCTION In database systems, data provenance =-=[4]-=- is a well-known concept, primarily used to answer questions concerning how query results are derived and which data sources they come from. A similar notion – network provenance [25] – is emerging in... |
219 | Adding structure to unstructured data
- Buneman, Davidson, et al.
- 1997
(Show Context)
Citation Context ...ging distributed systems [22, 14] and for accepting/rejecting network packets based on their traversed path. Alternatively, provenance may be more compactly represented using algebraic representation =-=[7, 3]-=-. Algebraic representations encode provenance using the binary operations + and ∗ (representing union and join, respectively). For instance, if α, β, and γ are the respective unique tuple IDs for link... |
210 | Implementing Declarative Overlays - LOO, CONDIE, et al. - 2005 |
132 | Declarative Routing: Extensible Routing with Declarative Queries
- LOO, HELLERSTEIN, et al.
- 2005
(Show Context)
Citation Context ...via a distributed query. Efficient provenance maintenance and querying. To maintain network provenance efficiently, we leverage the distributed query processing capabilities of declarative networking =-=[17, 16, 15]-=-. Declarative networking models network protocols as continuous queries over distributed streams. Declarative networking programs permit a variety of distributed network protocols to be specified conc... |
120 | Declarative networking: language, execution and optimization.
- Loo, Condie, et al.
- 2006
(Show Context)
Citation Context ...via a distributed query. Efficient provenance maintenance and querying. To maintain network provenance efficiently, we leverage the distributed query processing capabilities of declarative networking =-=[17, 16, 15]-=-. Declarative networking models network protocols as continuous queries over distributed streams. Declarative networking programs permit a variety of distributed network protocols to be specified conc... |
119 | A Survey of Research on Deductive Database Systems,
- Ramakrishnan, Ullman
- 1993
(Show Context)
Citation Context ...ative statements, declarative networking represents a natural means for unifying the synthesis and analysis of distributed protocols. The declarative NDlog language used by ExSPAN is based on Datalog =-=[20]-=-. A Datalog program consists of a set of rules. Each rule has the form p :- q1, q2, ..., qn., which can be read informally as “q1 and q2 and ... and qn imply p”. Here, p is the head of the rule, and q... |
48 | Building a Time Machine for Efficient Recording and Retrieval of High-Volume Network Traffic. In:
- Kornexl, Paxson, et al.
- 2005
(Show Context)
Citation Context ...rming network diagnostics, identifying malicious and misbehaving users, and enforcing trust management policies in distributed systems. Each goal has led to a series of application-specific proposals =-=[21, 1, 11, 24, 9, 12]-=- that focus on improving network support for accountability and providing efficient mechanisms to trace packets and information flows through the Internet. This paper explores the generic data managem... |
39 | Z.: D3s: debugging deployed distributed systems - Liu, Guo, et al. - 2008 |
38 | Providing Packet Obituaries,” in
- Argyraki, Maniatis, et al.
- 2004
(Show Context)
Citation Context ...rming network diagnostics, identifying malicious and misbehaving users, and enforcing trust management policies in distributed systems. Each goal has led to a series of application-specific proposals =-=[21, 1, 11, 24, 9, 12]-=- that focus on improving network support for accountability and providing efficient mechanisms to trace packets and information flows through the Internet. This paper explores the generic data managem... |
35 | Network monitors and contracting systems: competition and innovation
- Laskowski, Chuang
- 2006
(Show Context)
Citation Context ...rming network diagnostics, identifying malicious and misbehaving users, and enforcing trust management policies in distributed systems. Each goal has led to a series of application-specific proposals =-=[21, 1, 11, 24, 9, 12]-=- that focus on improving network support for accountability and providing efficient mechanisms to trace packets and information flows through the Internet. This paper explores the generic data managem... |
34 | Orchestra: rapid, collaborative sharing of dynamic data, in: CIDR,
- Ives, Khandelwal, et al.
- 2005
(Show Context)
Citation Context ...ta model for network provenance. We define a distributed datamodel for storing network provenance. Our data model builds upon current work on representing provenance information as relational tables =-=[10, 5]-=-, with extensions to supported distributed storage and querying. We propose two forms of distribution: a value-based approach in which all relevant information is piggy-backed onto communicated tuples... |
32 | Perm: Processing provenance and data on the same data model through query rewriting
- Glavic, Alonso
- 2009
(Show Context)
Citation Context ...ta model for network provenance. We define a distributed datamodel for storing network provenance. Our data model builds upon current work on representing provenance information as relational tables =-=[10, 5]-=-, with extensions to supported distributed storage and querying. We propose two forms of distribution: a value-based approach in which all relevant information is piggy-backed onto communicated tuples... |
27 | Forensic analysis for epidemic attacks in federated networks.
- Xie, Sekar, et al.
- 2006
(Show Context)
Citation Context ...rming network diagnostics, identifying malicious and misbehaving users, and enforcing trust management policies in distributed systems. Each goal has led to a series of application-specific proposals =-=[21, 1, 11, 24, 9, 12]-=- that focus on improving network support for accountability and providing efficient mechanisms to trace packets and information flows through the Internet. This paper explores the generic data managem... |
22 | PlanetFlow: Maintaining Accountability for Network Services.
- Huang, Bavier, et al.
- 2006
(Show Context)
Citation Context ...rming network diagnostics, identifying malicious and misbehaving users, and enforcing trust management policies in distributed systems. Each goal has led to a series of application-specific proposals =-=[21, 1, 11, 24, 9, 12]-=- that focus on improving network support for accountability and providing efficient mechanisms to trace packets and information flows through the Internet. This paper explores the generic data managem... |
13 |
Declarative toolkit for rapid network protocol simulation and experimentation
- MUTHUKUMAR, LI, et al.
- 2009
(Show Context)
Citation Context ... techniques aimed at reducing communication latency and bandwidth utilization. ExSPAN prototype implementation and evaluation. We present the prototype of ExSPAN. Our implementation utilizes RapidNet =-=[18]-=-, a declarative networking platform developed using the ns-3 network simulator [19]. Our experiments over simulated networks and an actual deployment on a testbed environment demonstrate that ExSPAN s... |
10 |
Distributed Monitoring and Forensics in Overlay Networks
- Singh, Maniatis, et al.
- 2006
(Show Context)
Citation Context ... tuple-level provenance information and is typically used to answer queries pertaining to fine-grained network state. For example, graph representation may be useful for debugging distributed systems =-=[22, 14]-=- and for accepting/rejecting network packets based on their traversed path. Alternatively, provenance may be more compactly represented using algebraic representation [7, 3]. Algebraic representations... |
7 |
Modelling Topology of Large Networks,” http:// www.cc.gatech.edu/projects/gtitm
- GT-ITM
- 2010
(Show Context)
Citation Context ...y the ns-3 runtime. ExSPAN uses the identical codebase for both simulation and deployment modes. We generate transit-stub topologies for our simulation experiments using the GT-ITM topology generator =-=[8]-=-. The transit-stub topology consists of eight nodes per stub, three stubs per transit node, and four nodes per transit domain. We increase the number of nodes in the network by increasing the number o... |