Scalable flow-based networking with DIFANE

By Minlan Yu, Jennifer Rexford, M. Freedman, Jia Wang. Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication, 2010

701 citations recorded by the source corpus.

Ideally, enterprise administrators could specify fine-grain policies that drive how the underlying switches forward, drop, and measure traffic. However, existing techniques for flow-based networking rely too heavily on centralized controller software that installs rules reactively, based on the first packet of each flow. In this paper, we propose DIFANE, a scalable and efficient solution that keeps all traffic in the data plane by selectively directing packets through intermediate switches that store the necessary rules. DIFANE relegates the controller to the simpler task of partitioning these rules over the switches. DIFANE can be readily implemented with commodity switch hardware, since all data-plane functions can be expressed in terms of wildcard rules that perform simple actions on matching packets. Experiments with our prototype on Click-based OpenFlow switches show that DIFANE scales to larger networks with richer policies.

Subfields: Computer Networks and Communications, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/6559538bb88b5d78fa14525dd0d374da3263edd6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Hardware and Architecture

Citations in the corpus

AVANT-GUARD: scalable and vigilant switch flow management in software-defined networks (2013, background): “However, the goals and main approach of De-voFlow and DIFANE are very different from our work in that these studies are focused on improving performance under normal network conditions.”
Optimizing the "one big switch" abstraction in software-defined networks (2013, background): “Distributing policy while changing routing: DI-FANE [19] and vCRIB [20] leverage all switches in the network to enforce an endpoint policy.”
Security in Software Defined Networks: A Survey (2015, positive): “There are some proposals that increase controller availability by minimizing controllers’ responsibility such as DevoFLow [148] and DIFANE [149].”
FloodGuard: A DoS Attack Prevention Extension in Software-Defined Networks (2015, methodology): “While there are many studies and solutions on the scalability issue [20], [12], [10], [32], [14], there is very little research on the even more challenging security issue.”
Maestro: A System for Scalable OpenFlow Control (2010, methodology): “DIFANE [20] provides a way to achieve efficient rulebased policy enforcement in a network by performing policy rules matching at the switches themselves.”
On Controller Performance in Software-Defined Networks (2012, methodology): “, [6, 18]) to address perceived architectural inefficiencies.”
A Survey of Software-Defined Networking: Past, Present, and Future of Programmable Networks (2014, background): “A good example of proactive control is DIFANE [35], which partitions rules over a hierarchy of switches, such that the controller rarely needs to be consulted about new ﬂows and trafﬁc is kept within the data-plane.”
Distributed SDN Control: Survey, Taxonomy, and Challenges (2018, methodology): “Contrary to DevoFlow [28] and DIFANE [27], Kandoo proposes to reduce the overall stress on the control plane without the need to modify OpenFlow switches.”
Software defined networking: State of the art and research challenges (2014, positive): “However, the DIFANE [54] architecture leverages these limitations by distributing the OpenFlow wildcard rules among various switches to ensure that forwarding decisions occur in the data plane.”
OFLOPS: An Open Framework for OpenFlow Switch Evaluation (2012, background): “OpenFlow is increasingly adopted, both by hardware vendors as well as by the research community [9,19,16].”
Survey on Network Virtualization Hypervisors for Software Defined Networking (2015, background): “Hypervisors that execute hypervisor functions on the data plane, i.e., the distributed hypervisors that involve NEs (see Sections V-B–V-D), require special consideration when evaluating control plane isolation.”
DISCO: Distributed multi-domain SDN controllers (2013, methodology): “A ﬁrst class of solutions, such as DIFANE [15] and DevoFlow [16], address this problem by extending data plane mechanisms of switches with the objective of reducing the load towards the controller.”
DevoFlow: scaling flow management for high-performance networks (2011, background): “DIFANE [51] avoids these overheads by splitting pre-installed OpenFlow wildcard rules among multiple switches, in a clever way that ensures all decisions can be made in the data-plane.”
A roadmap for traffic engineering in SDN-OpenFlow networks (2014, methodology): “DIFANE [41] Distributed controller’s rules across a subset of the authority switches.”
SIMPLE-fying middlebox policy enforcement using SDN (2013, background): “Due to space constraints, we do not show the formal hardness reduction; at a high-level the intractability is due to the integer constraints necessary to model the switch table sizes.”
Software Defined Networking Architecture, Security and Energy Efficiency: A Survey (2017, methodology): “An approach called DIFANE has been capable of achieving controller load balancing by implementing detailed and strict policies in enterprise networks [143].”