Results 1 - 10
of
72
Onix: a distributed control platform for large-scale production networks.
- In USENIX OSDI,
, 2010
"... Abstract Computer networks lack a general control paradigm, as traditional networks do not provide any networkwide management abstractions. As a result, each new function (such as routing) must provide its own state distribution, element discovery, and failure recovery mechanisms. We believe this l ..."
Abstract
-
Cited by 164 (10 self)
- Add to MetaCart
(Show Context)
Abstract Computer networks lack a general control paradigm, as traditional networks do not provide any networkwide management abstractions. As a result, each new function (such as routing) must provide its own state distribution, element discovery, and failure recovery mechanisms. We believe this lack of a common control platform has significantly hindered the development of flexible, reliable and feature-rich network control planes. To address this, we present Onix, a platform on top of which a network control plane can be implemented as a distributed system. Control planes written within Onix operate on a global view of the network, and use basic state distribution primitives provided by the platform. Thus Onix provides a general API for control plane implementations, while allowing them to make their own trade-offs among consistency, durability, and scalability.
Devoflow: Scaling flow management for high-performance networks
- In ACM SIGCOMM
, 2011
"... OpenFlow is a great concept, but its original design imposes excessive overheads. It can simplify network and traffic management in enterprise and data center environments, because it enables flow-level control over Ethernet switching and provides global visibility of the flows in the network. Howev ..."
Abstract
-
Cited by 134 (1 self)
- Add to MetaCart
(Show Context)
OpenFlow is a great concept, but its original design imposes excessive overheads. It can simplify network and traffic management in enterprise and data center environments, because it enables flow-level control over Ethernet switching and provides global visibility of the flows in the network. However, such fine-grained control and visibility comes with costs: the switch-implementation costs of involving the switch’s control-plane too often and the distributed-system costs of involving the OpenFlow controller too frequently, both on flow setups and especially for statistics-gathering. In this paper, we analyze these overheads, and show that OpenFlow’s current design cannot meet the needs of highperformance networks. We design and evaluate DevoFlow, a modification of the OpenFlow model which gently breaks the coupling between control and global visibility, in a way that maintains a useful amount of visibility without imposing unnecessary costs. We evaluate DevoFlow through simulations, and find that it can load-balance data center traffic as well as fine-grained solutions, without as much overhead: DevoFlow uses 10–53 times fewer flow table entries at an average switch, and uses 10–42 times fewer control messages.
VeriFlow: Verifying networkwide invariants in real time
- In HotSDN
, 2012
"... Networks are complex and prone to bugs. Existing tools that check network configuration files and the data-plane state operate offline at timescales of seconds to hours, and cannot detect or prevent bugs as they arise. Is it possible to check network-wide invariants in real time, as the network stat ..."
Abstract
-
Cited by 80 (3 self)
- Add to MetaCart
(Show Context)
Networks are complex and prone to bugs. Existing tools that check network configuration files and the data-plane state operate offline at timescales of seconds to hours, and cannot detect or prevent bugs as they arise. Is it possible to check network-wide invariants in real time, as the network state evolves? The key challenge here is to achieve extremely low latency during the checks so that network performance is not affected. In this paper, we present a design, VeriFlow, which achieves this goal. VeriFlow is a layer between a softwaredefined networking controller and network devices that checks for network-wide invariant violations dynamically as each forwarding rule is inserted, modified or deleted. VeriFlow supports analysis over multiple header fields, and an API for checking custom invariants. Based on a prototype implementation integrated with the NOX OpenFlow controller, and driven by a Mininet OpenFlow network and Route Views trace data, we find that Veri-Flow can perform rigorous checking within hundreds of microseconds per rule insertion or deletion. 1
OFLOPS: An Open Framework for Openflow Switch Evaluation,” in PAM,
, 2012
"... Abstract. Recent efforts in software-defined networks, such as OpenFlow, give unprecedented access into the forwarding plane of networking equipment. When building a network based on OpenFlow however, one must take into account the performance characteristics of particular OpenFlow switch implement ..."
Abstract
-
Cited by 49 (4 self)
- Add to MetaCart
(Show Context)
Abstract. Recent efforts in software-defined networks, such as OpenFlow, give unprecedented access into the forwarding plane of networking equipment. When building a network based on OpenFlow however, one must take into account the performance characteristics of particular OpenFlow switch implementations. In this paper, we present OFLOPS, an open and generic software framework that permits the development of tests for OpenFlow-enabled switches, that measure the capabilities and bottlenecks between the forwarding engine of the switch and the remote control application. OFLOPS combines hardware instrumentation with an extensible software framework. We use OFLOPS to evaluate current OpenFlow switch implementations and make the following observations: (i) The switching performance of flows depends on applied actions and firmware. (ii) Current OpenFlow implementations differ substantially in flow updating rates as well as traffic monitoring capabilities. (iii) Accurate OpenFlow command completion can be observed only through the data plane. These observations are crucial for understanding the applicability of OpenFlow in the context of specific use-cases, which have requirements in terms of forwarding table consistency, flow setup latency, flow space granularity, packet modification types, and/or traffic monitoring abilities.
Composing software defined networks
- in NSDI
, 2013
"... In Software Defined Networking (SDN), an application comprising many disparate tasks must be converted to a single set of packet-processing rules on the switches. Unfortunately, today’s SDN platforms do not support expressing these tasks as separate modules, and composing them to create an applicati ..."
Abstract
-
Cited by 40 (7 self)
- Add to MetaCart
(Show Context)
In Software Defined Networking (SDN), an application comprising many disparate tasks must be converted to a single set of packet-processing rules on the switches. Unfortunately, today’s SDN platforms do not support expressing these tasks as separate modules, and composing them to create an application. This leads to monolithic programs that are neither portable, nor reusable. In this paper, we present the FV system that presents each module with an abstract view of the network topology customized to the application logic, where one module may implement the “switching fabric ” for another. For example, a firewall module may run on “one big switch ” that is implemented by a routing module. The programmer can specify network views, as well as the relationship between (virtual) switches in different views. For example, conceptually the firewall functionality runs before the routing functionality. Using sequential composition, the FV compiler can synthesize a single set of rules and queries for each physical switch. FV includes a core language for defining policies as mathematical functions in an imperative style familiar to Python programmers, as well as a module language that supports abstraction (i.e., network views) and protection (i.e., specifying what traffic a module can measure and control). FV enables the creation of sophisticated SDN applications, as illustrated by example programs running on our FV prototype. 1
XIA: Efficient Support for Evolvable Internetworking
"... Motivated by limitations in today’s host-centric IP network, recent studies have proposed clean-slate network architectures centered around alternate first-class principals, such as content, services, or users. However, much like the host-centric IP design, elevating one principal type above others ..."
Abstract
-
Cited by 31 (8 self)
- Add to MetaCart
(Show Context)
Motivated by limitations in today’s host-centric IP network, recent studies have proposed clean-slate network architectures centered around alternate first-class principals, such as content, services, or users. However, much like the host-centric IP design, elevating one principal type above others hinders communication between other principals and inhibits the network’s capability to evolve. This paper presents the eXpressive Internet Architecture (XIA), an architecture with native support for multiple principals and the ability to evolve its functionality to accommodate new, as yet unforeseen, principals over time. We describe key design requirements, and demonstrate how XIA’s rich addressing and forwarding semantics facilitate flexibility and evolvability, while keeping core network functions simple and efficient. We describe case studies that demonstrate key functionality XIA enables. 1
NetLord: A Scalable Multi-Tenant Network Architecture for
"... Providers of “Infrastructure-as-a-Service ” need datacenter networks that support multi-tenancy, scale, and ease of operation, at low cost. Most existing network architectures cannot meet all of these needs simultaneously. In this paper we present NetLord, a novel multi-tenant network architecture. ..."
Abstract
-
Cited by 26 (0 self)
- Add to MetaCart
(Show Context)
Providers of “Infrastructure-as-a-Service ” need datacenter networks that support multi-tenancy, scale, and ease of operation, at low cost. Most existing network architectures cannot meet all of these needs simultaneously. In this paper we present NetLord, a novel multi-tenant network architecture. NetLord provides tenants with simple and flexible network abstractions, by fully and efficiently virtualizing the address space at both L2 and L3. NetLord can exploit inexpensive commodity equipment to scale the network to several thousands of tenants and millions of virtual machines. NetLord requires only a small amount of offline, one-time configuration. We implemented NetLord on a testbed, and demonstrated its scalability, while achieving order-of-magnitude goodput improvements over previous approaches.
SoftCell: Scalable and Flexible Cellular Core Network Architecture
"... Existing cellular networks suffer from inflexible and expensive equipment, and complex control-plane protocols. To address these challenges, we present SoftCell, a scalable architecture that supports fine-grained policies for mobile devices in cellular core networks, using commodity switches and ser ..."
Abstract
-
Cited by 24 (2 self)
- Add to MetaCart
(Show Context)
Existing cellular networks suffer from inflexible and expensive equipment, and complex control-plane protocols. To address these challenges, we present SoftCell, a scalable architecture that supports fine-grained policies for mobile devices in cellular core networks, using commodity switches and servers. A controller realizes high-level service polices that direct traffic through sequences of middleboxes based on subscriber attributes and applications. To enable small forwarding tables in core switches, SoftCell aggregates traffic along multiple dimensions—the service policy, the base station, and the mobile device—at different switches in the network. Since most traffic originates from mobile devices, SoftCell performs fine-grain packet classification at the access switches at the base stations, where software switches can easily handle the state and bandwidth requirements. Soft-Cell guarantees that packets in the same connection traverse the same sequence of middleboxes in both directions, even in the presence of mobility, without requiring expensive packet classification at the high-bandwidth gateway edge switches. We demonstrate that SoftCell improves the scalability and flexibility of cellular core networks through analysis of LTE workloads, micro-benchmarks on our prototype controller, and large-scale simulations. 1.
Slicing home networks
- In Proceedings of the 2nd ACM SIGCOMM workshop on Home networks (2011), ACM
"... Despite the popularity of home networks, they face a number of systemic problems: (i) Broadband networks are expensive to deploy; and it is not clear how the cost can be shared by several service providers; (ii) Home networks are getting harder to manage as we connect more devices, use new applicati ..."
Abstract
-
Cited by 23 (2 self)
- Add to MetaCart
(Show Context)
Despite the popularity of home networks, they face a number of systemic problems: (i) Broadband networks are expensive to deploy; and it is not clear how the cost can be shared by several service providers; (ii) Home networks are getting harder to manage as we connect more devices, use new applications, and rely on them for entertainment, communication and work—it is common for home networks to be poorly managed, insecure or just plain broken; and (iii) It is not clear how home networks will steadily improve, after they have been deployed, to provide steadily better service to home users. In this paper we propose slicing home networks as a way to overcome these problems. As a mechanism, slicing allows multiple service providers to share a common infrastructure; and supports many policies and business models for cost sharing. We propose four requirements for slicing home networks: bandwidth and traffic isolation between slices, independent control of each slice, and the ability to modify and improve the behavior of a slice. We explore how these requirements allow cost-sharing, out-sourced management of home networks, and the ability to customize a slice to provide higher-quality service. Finally, we describe an initial prototype that we are deploying in homes.
Software-Defined Networking: A Comprehensive Survey
, 2014
"... The Internet has led to the creation of a digital society, where (almost) everything is connected and is accessible from anywhere. However, despite their widespread adoption, traditional IP networks are complex and very hard to manage. It is both difficult to configure the network according to pre- ..."
Abstract
-
Cited by 20 (3 self)
- Add to MetaCart
The Internet has led to the creation of a digital society, where (almost) everything is connected and is accessible from anywhere. However, despite their widespread adoption, traditional IP networks are complex and very hard to manage. It is both difficult to configure the network according to pre-defined policies, and to reconfigure it to respond to faults, load and changes. To make matters even more difficult, current networks are also vertically integrated: the control and data planes are bundled together. Software-Defined Networking (SDN) is an emerging paradigm that promises to change this state of affairs, by breaking vertical integration, separating the network’s control logic from the underlying routers and switches, promoting (logical) centralization of network control, and introducing the ability to program the network. The separation of concerns introduced between the definition of network policies, their
