Results 1 - 10
of
22
Online strategies for intra and inter provider service migration in virtual networks
- IN: PROC. PRINCIPLES, SYSTEMS AND APPLICATIONS OF IP TELECOMMUNICATIONS, IPTCOMM
, 2011
"... Network virtualization allows one to build dynamic distributed systems in which resources can be dynamically allocated at locations where they are most useful. In order to fully exploit the benefits of this new technology, protocols need to be devised which react efficiently to changes in the deman ..."
Abstract
-
Cited by 10 (7 self)
- Add to MetaCart
(Show Context)
Network virtualization allows one to build dynamic distributed systems in which resources can be dynamically allocated at locations where they are most useful. In order to fully exploit the benefits of this new technology, protocols need to be devised which react efficiently to changes in the demand. This paper argues that the field of online algorithms and competitive analysis provides useful tools to deal with and reason about the uncertainty in the request dynamics, and to design algorithms with provable performance guarantees. As a case study, we describe a system (e.g., a gaming application) where network virtualization is used to support thin client applications for mobile devices to improve their QoS. By decoupling the service from the underlying resource infrastructure, it can be migrated closer to the current client locations while taking into account migration cost. This paper identifies the major cost factors in such a system, and formalizes the corresponding optimization problem. Both randomized and deterministic, gravity center based online algorithms are presented which achieve a good tradeoff between improved QoS and migration cost in the worst-case, both for service migration within an infrastructure provider as well as for networks supporting cross-provider migration. The paper reports on our simulation results and also presents an explicit construction of an optimal offline algorithm which allows, e.g., to evaluate the competitive ratio empirically.
An Opportunistic Resource Sharing and Topology-Aware Mapping Framework for Virtual Networks
"... Network virtualization provides a promising way to overcome Internet ossification. A major challenge is virtual network mapping, i.e., how to embed multiple virtual network requests with resource constraints into a substrate network, such that physical resources are utilized in an efficient and eff ..."
Abstract
-
Cited by 8 (2 self)
- Add to MetaCart
Network virtualization provides a promising way to overcome Internet ossification. A major challenge is virtual network mapping, i.e., how to embed multiple virtual network requests with resource constraints into a substrate network, such that physical resources are utilized in an efficient and effective manner. Since this problem is known to be NP-complete, a variety of heuristic algorithms have been proposed. In this paper, we re-examine this problem and propose a virtual network mapping framework, ORSTA, which is based on Opportunistic Resource Sharing and Topology-Aware node ranking. Opportunistic re-source sharing is taken into consideration at the entire network level for the first time and we develop an online approxima-tion algorithm, FFA, for solving the corresponding time slot assignment problem. To measure the topology importance of a substrate node, a node ranking method, MCRank, based on Markov chain is presented. We also devise a simple and practical method to estimate the residual resource of a substrate node/link. Extensive simulation experiments demonstrate that the proposed framework enables the substrate network to achieve efficient physical resource utilization and to accept many more virtual network requests over time.
A General Distributed Approach to Slice Embedding with Guarantees
"... The challenge of deploying and managing virtualization based network services (slices) recently spurred interest in both the business and the research communities. To provide wide-area network services, resources from different infrastructure providers are needed. Leveraging the consensusbased task ..."
Abstract
-
Cited by 6 (6 self)
- Add to MetaCart
The challenge of deploying and managing virtualization based network services (slices) recently spurred interest in both the business and the research communities. To provide wide-area network services, resources from different infrastructure providers are needed. Leveraging the consensusbased task allocation algorithms from the robotics literature, we propose a general distributed auction mechanism for the (NPhard) slice embedding problem. Under reasonable assumptions on the bidding scheme, the proposed mechanism is proven to converge, and it is shown that the solutions guarantee a worstcase efficiency of (1 − 1) w.r.t. the optimal. Using extensive e simulations, we confirm superior convergence properties and resource utilization when compared with existing distributed slice embedding solutions, and we show how by appropriate policy design, our mechanism can be instantiated to accommodate the embedding goals of different service and infrastructure providers.
NetworkVirtualization Nodes that Support Mutually Independent Development and Evolution of Components
"... Abstract – “Virtualization nodes ” (VNodes) for programmable network-virtualization platforms are being developed. Criteria for “clean ” network-virtualization are devised and applied to this platform and slices (virtual networks). These criteria meet one of the challenges targeted by the Virtualiza ..."
Abstract
-
Cited by 5 (5 self)
- Add to MetaCart
Abstract – “Virtualization nodes ” (VNodes) for programmable network-virtualization platforms are being developed. Criteria for “clean ” network-virtualization are devised and applied to this platform and slices (virtual networks). These criteria meet one of the challenges targeted by the Virtualization Node Project, that is, to enable mutually independent development and evolution of components (namely, computational components called programmers and networking components called redirectors) in VNodes. To meet this challenge, the redirector plays the central role in implementing the following two functions of VNodes. The first function is creation of mapping between virtual links to external physical paths and mapping between virtual links to internal physical paths, which makes it possible to hide various alternative computational components in the VNode from the external network and to hide these external-network representations from the internal components. The second function is implementation of highperformance data conversion, which connects the external and internal data formats or mappings, by using an add-on card with a network processor. Two results are obtained from the performance evaluation of these functions. First, the overhead caused by mapping creation can be hidden by other tasks under normal conditions, but the overhead caused by mapping deletion must be reduced. Second, the data-conversion rate is half the wire rate, which should be increased in future work. I.
Virtual Network Embedding with Opportunistic Resource Sharing
"... Network virtualization has emerged as a promising approach to overcome the ossification of the Internet. A major challenge in network virtualization is the so-called virtual network embedding problem, which deals with the efficient embedding of virtual networks with resource constraints into a shar ..."
Abstract
-
Cited by 3 (1 self)
- Add to MetaCart
Network virtualization has emerged as a promising approach to overcome the ossification of the Internet. A major challenge in network virtualization is the so-called virtual network embedding problem, which deals with the efficient embedding of virtual networks with resource constraints into a shared substrate network. A number of heuristics have been proposed to cope with the NP-hardness of this problem; however, all of the existing proposals reserve fixed resources throughout the entire lifetime of a virtual network. In this paper, we re-examine this problem with the position that time-varying resource requirements of virtual networks should be taken into consideration, and we present an opportunistic resource sharing-based mapping framework, ORS, where substrate resources are opportunistically shared among multiple virtual networks. We formulate the time slot assignment as an optimization problem, then we prove the decision version of the problem to be NP-hard in the strong sense. Observing the resemblance between our problem and the bin packing problem, we adopt the core idea of first-fit, and propose two practical solutions: first-fit by collision probability (CFF) and first-fit by expectation of indicators’ sum (EFF). Simulation results show that that ORS provides a more efficient utilization of substrate resources than two state-of-the-art fixed-resource embedding schemes.
Progressive Virtual Topology Embedding in OpenFlow Networks
"... Abstract—Future internet would provide a flexible and simpler architectural design by combining novel low level clean–slate techniques such as OpenFlow with high level design princi-ples such as network virtualization. However, previous research outlined that at the core of network virtualization st ..."
Abstract
-
Cited by 2 (1 self)
- Add to MetaCart
(Show Context)
Abstract—Future internet would provide a flexible and simpler architectural design by combining novel low level clean–slate techniques such as OpenFlow with high level design princi-ples such as network virtualization. However, previous research outlined that at the core of network virtualization stands a new set of challenges for network resources allocation. In this work we focus on one such challenge, namely the problem of virtual topology embedding. In this context users need to leverage the infrastructure substrate by accomodating logical topologies with high degree of flexibility. The network provider, on the other hand, aims at maximizing its revenue in term of size and number of topologies accepted while minimizing costs accounting for the substrate network resources used. To this aim, we present VT-Planner a novel virtual network embedding technique with reduced computational cost and very efficient over substrate topologies encountered in practice. Extensive numerical simulations are provided comparing this technique with state– of–the–art solutions: our results show that VT-Planner is able to achieve a good balance in terms of complexity and performance. Index Terms—Resource allocation, Virtual Topology Embed-ding, OpenFlow, Algorithms
A Decomposition-based Architecture for Distributed Virtual Network Embedding
"... Network protocols have historically been developed on an ad-hoc basis, and cloud computing is no exception. A fundamental management protocol, not yet standardized, that cloud providers need to run to support wide-area virtual network services is the virtual network (VN) embedding protocol. In thi ..."
Abstract
-
Cited by 2 (2 self)
- Add to MetaCart
(Show Context)
Network protocols have historically been developed on an ad-hoc basis, and cloud computing is no exception. A fundamental management protocol, not yet standardized, that cloud providers need to run to support wide-area virtual network services is the virtual network (VN) embedding protocol. In this paper, we use decomposition theory to provide a unifying architecture for the VN embedding problem. We show how our architecture subsumes existing solutions, and how it can be used by cloud providers to design a distributed VN embedding protocol that adapts to different scenarios, by merely instantiating different decomposition policies. We analyze key representative tradeoffs via simulation, and with our VN embedding testbed that uses a Linux system architecture to reserve virtual node and link capacities. In contrast with existing VN embedding solutions, we found that partitioning a VN request not only increases the signaling overhead, but may decrease cloud providers’ revenue.
Virtual network embedding with substrate support for parallelization, in
- Proc. of IEEE Global Communication Conference (GLOBECOM
, 2012
"... Abstract—Network virtualization has been the focus of intense research interest and is a promising approach to overcome the ossification of the Internet. A major challenge with network virtualization is virtual network embedding, which deals with the efficient embedding of virtual networks with reso ..."
Abstract
-
Cited by 2 (1 self)
- Add to MetaCart
(Show Context)
Abstract—Network virtualization has been the focus of intense research interest and is a promising approach to overcome the ossification of the Internet. A major challenge with network virtualization is virtual network embedding, which deals with the efficient embedding of virtual networks with resource constraints into a substrate network. Many research results have been reported regarding this problem. However, there hasn’t been any focus on virtual network embedding with substrate support for parallelization, i.e., the substrate network supports parallel computation and allows a virtual node to be mapped into multiple substrate nodes. This paper is the first attempt at gaining a better understanding on how parallelization benefits embedding. We present a formal problem description and propose two algorithms that capitalize parallelism. Several extensions are developed to complement the proposed algorithms. From experimental results, the effectiveness and usefulness of the algorithms and extensions are confirmed. I.
Specificity vs. Flexibility: On the Embedding Cost of a Virtual Network
"... Abstract—The virtualization trend in today’s Internet decouples services from the constraints of the underlying physical infrastructure. This decoupling facilitates more flexible and efficient resource allocations: the service can be realized at any place in the substrate network which fulfills the ..."
Abstract
-
Cited by 1 (1 self)
- Add to MetaCart
(Show Context)
Abstract—The virtualization trend in today’s Internet decouples services from the constraints of the underlying physical infrastructure. This decoupling facilitates more flexible and efficient resource allocations: the service can be realized at any place in the substrate network which fulfills the service specification requirements. This paper studies such flexibilities in the context of virtual network (VNet) embeddings. The network virtualization paradigm envisions an Internet where users can request arbitrary VNets from a substrate provider (e.g., an ISP). A VNet describes a set of virtual nodes which are connected by virtual links; both nodes and links provide certain QoS or resource guarantees. While some parts of the VNet may be fully specified (e.g., the node and link locations or technologies), other parts may be flexible or left open entirely. We analyze how flexible specifications can be exploited to improve the embedding of virtual networks. We define a measure for specificity and introduce the notion of the Price of Specificity (PoS) which captures the resource cost of the embedding under a given specification. We identify parameters on which the Price of Specificity depends, and evaluate its magnitude in different scenarios. For example, we find that the PoS can be large even in small settings, and depends both on the substrate size as well as—to a larger extent—the load of the substrate. Moreover, while skewed distributions of resources can yield better allocations, they entail the risks of a high PoS if the demand does not perfectly match. We also provide a formal analysis of the impact of migration, and show that the option to migrate can sometimes increase resource costs. I.
The price of specificity in the age of network virtualization
- In Proc. 5th IEEE/ACM UCC
, 2012
"... Abstract—The virtualization trend in today’s Internet decouples services from the constraints of the underlying physical infrastructure. This decoupling has the potential to facilitate more flexible and efficient resource allocations: the service can be realized at any place in the substrate network ..."
Abstract
-
Cited by 1 (1 self)
- Add to MetaCart
(Show Context)
Abstract—The virtualization trend in today’s Internet decouples services from the constraints of the underlying physical infrastructure. This decoupling has the potential to facilitate more flexible and efficient resource allocations: the service can be realized at any place in the substrate network which fulfills the service specification requirements. This paper studies such flexibilities in the context of virtual network (VNet) embeddings. The network virtualization paradigm envisions an Internet where users can request arbitrary VNets from a substrate provider (e.g., an ISP). A VNet describes a set of virtual nodes which are connected by virtual links; both nodes and links provide certain resource guarantees. While some parts of the VNet may be fully specified (e.g., the node and link locations or technologies), other parts may be flexible or left open entirely. For example, it may be irrelevant for some users on which vendor hardware the VNet is realized, or a user only requests that the VNet runs in some European cloud provider. We study how flexible specifications can be exploited to improve the embedding of virtual networks. We introduce the notion of the Price of Specificity which captures the resource cost of the embedding under a given specification. We analyze on which parameters the Price of Specificity depends, and evaluate its magnitude in different scenarios. I.
