Multi-Provider Service Chain Embedding With Nestor

Abstract

Abstract-Network Function Virtualization (NFV) decouples network functions (NF) from the underlying middlebox hardware and promotes their deployment on virtualized network infrastructures. This essentially paves the way for the migration of NFs into clouds (i.e., NF-as-a-Service), achieving a drastic reduction of middlebox investment and operational costs for enterprises. In this context, service chains (expressing middlebox policies in the enterprise network) should be mapped onto datacenter networks, ensuring correctness, resource efficiency as well as compliance with the provider's policy. The network service embedding (NSE) problem is further exacerbated by two challenging aspects: (i) traffic scaling caused by certain NFs (e.g., caches, WAN optimizers) and (ii) NF location dependencies. Traffic scaling requires resource reservations different from the ones specified in the service chain, whereas NF location dependencies, in conjunction with the limited geographic footprint of NF providers (NFPs), raise the need for NSE across multiple NFPs.In this paper, we present a holistic solution to the multiprovider NSE problem. We decompose NSE into (i) NF-graph partitioning performed by a centralized coordinator and (ii) NF-subgraph mapping onto datacenter networks. We present linear programming formulations to derive near-optimal solutions for both problems. We address the challenging aspect of traffic scaling by introducing a new service model that supports demand transformations. We also define topology abstractions for NF-graph partitioning. Furthermore, we discuss the steps required to embed service chains across multiple NFPs, using our NSE orchestrator (Nestor). We perform an evaluation study of multi-provider NSE with emphasis on NF-graph partitioning optimizations tailored to the client and NFPs. Our evaluation results further uncover significant savings in terms of service cost and resource consumption due to the demand transformations.

Optimization Model for Cross-Domain Network Slices in 5G Networks

“…Dietrich et al [17] presented a holistic solution to the multi-provider network service embedding (NSE) problem to allow SFC mapping across multiple domains. They explained that the traffic scaling and NF location dependencies are the main challenging aspects of the problem.…”

A deep reinforcement learning-based algorithm for reliability-aware multi-domain service deployment in smart ecosystems

“…1 shows an IoT system in which information from the medical and traffic networks is relayed to the core data-center (with possibility that there are multiple core cloud servers with each belonging to different InPs) through an access network (wireless and optical) comprised of multiple InPs. Moreover, the location dependencies of certain network functions, coupled with the limited footprint of InPs may necessitate the deployment of SFCs of such services across substrate infrastructure belonging to multiple providers [14], [15], [16]. In this case, the end-to-end service is realized as a concatenation of service instances supported by different InPs.…”

“…The problem of multi-domain service deployment has been addressed in [25], [1], [26], [27], [28], [14], [29]. However, such heuristic approaches are not well suited for scenarios in which the service deployment decision requires to jointly consider multiple network attributes such as cost, delay and reliability among others.…”

Disaggregated optical network control and orchestration of heterogeneous domains

“…In turn, the interconnection network includes two main technologies: packet-based devices (e.g., SDN switches) and optical devices (e.g., ROADMs and optical transponders). In such a scenario, the orchestration process is required to address the application demands by coordinately provision a composite set of services across different administrative and/or technological domains (e.g., cloud and network domains, different service provider domains, packet-based and optical network domains), while automating the tasks of composing and exposing them as a single service instance [2].…”