Orchestration Algorithm Research Articles

Computation Offloading in Beyond 5G Networks: A Distributed Learning Framework and Applications

Facing the trend of merging wireless communications and multi-access edge computing (MEC), this article studies computation offloading in beyond fifth generation networks. To address the technical challenges originating from the uncertainties and the sharing of limited resource in an MEC system, we formulate the computation offloading problem as a multi-agent Markov decision process, for which a distributed learning framework is proposed. We present a case study on resource orchestration in computation offloading to showcase the potential of an online distributed reinforcement learning algorithm developed under the proposed framework. Experimental results demonstrate that our learning algorithm outperforms the benchmark resource orchestration algorithms. Furthermore, we outline the research directions worth in-depth investigation to minimize the time cost, which is one of the main practical issues that prevent the implementation of the proposed distributed learning framework.

Endogenous Trusted DRL-Based Service Function Chain Orchestration for IoT

With the development of the Internet of Things, trust has become a limited factor in the integration of heterogeneous IoT networks. In this regard, we use the combination of blockchain technology and SDN/NFV to build a heterogeneous IoT network resource management model based on the consortium chain. In order to solve the efficiency problem caused by the full amount of data on the chain, we deploy light nodes and full nodes for the consortium chain. At the same time, we use the idea of identification to realize the separation of identification and resource information, build the application mode of on-chain identification and off-chain information, and realize resources endogenous trust management. We also propose a practical Byzantine fault-tolerant consensus mechanism based on reputation value to save consensus costs and improve efficiency. Combined with artificial intelligence technology, we introduce deep reinforcement learning for service function chain orchestration, and design a service function chain orchestration algorithm based on Asynchronous Advantage Actor-Critic to optimize orchestration costs. The final simulation results show that the consensus algorithm and service function chain orchestration algorithm we designed have good performance in terms of cost saving and efficiency improvement.

PureEdgeSim: A simulation framework for performance evaluation of cloud, edge and mist computing environments

Edge and Mist Computing are two emerging paradigms that aim to reduce latency and the Cloud workload by bringing its applications close to the Internet of Things (IoT) devices. In such complex environments, simulation makes it possible to evaluate the adopted strategies before their deployment on a real distributed system. However, despite the research advancement in this area, simulation tools are lacking, especially in the case of Mist Computing [11], where heterogeneous and constrained devices cooperate and share their resources. Motivated by this, in this paper, we present PureEdgeSim, a simulation toolkit that enables the simulation of Cloud, Edge, and Mist Computing environments and the evaluation of the adopted resources management strategies, in terms of delays, energy consumption, resources utilization, and tasks success rate. To show its capabilities, we introduce a case study, in which we evaluate the different architectures, orchestration algorithms, and the impact of offloading criteria. The simulation results show the effectiveness of PureEdgeSim in modeling such complex and dynamic environments.

Profit Maximization of Online Service Function Chain Orchestration in an Inter-Datacenter Elastic Optical Network

To meet the increasing traffic demands characterized by large bandwidth and high burstiness, more traffic has been moving to inter-datacenter elastic optical networks (inter-DC EONs) for processing. The integration of two emerging paradigms, network function virtualization (NFV) and software-defined networking (SDN), enables Internet service providers (ISPs) to deploy service function chains (SFCs) from users flexibly while reducing operational and capital expenditures. This article focuses on the problem of online SFC provisioning in inter-DC-EONs with the aim of maximizing ISP profits, where the challenge in jointly allocating IT and spectrum resources when deploying SFCs is balanced with the deployment costs of processing as many user requests as possible. We design two-phase time-efficient orchestration algorithms for online SFC requests and the strategy of SFC splitting is adopted to improve the utilization of spectrum resources on fiber links. Simulation results show that, compared with the existing algorithm, our proposed algorithms significantly shorten the deployment time, improve total profit of ISP by up to 40% and reduce the blocking probability by up to 35%.

Virtual Network Functions Migration Cost: from Identification to Prediction

The advent of the function virtualization concept, especially that of network functions, leads to important benefits for future networks. Although the orchestration of virtualized functions presents gains for network operators and clients alike, the overhead for moving functions has not been thoroughly explored so far, especially considering functions virtualized by using container technologies. In this work, we investigate orchestration costs associated with the migration of containerized virtual functions. To this end, we first perform a systematic literature review on state-of-the-art virtual function migration costs, electing time and data transferred as so. We then use a well-known container platform (LXD) to perform several orchestration experiments in a controlled environment. By analyzing the container migration process in smaller complementary steps, and designing experiments to evaluate them individually, a pattern for migration costs is observed. Linear regression is then used to derive a prediction model for the necessary time and data transferring for performing a container migration. To assess the predictor’s accuracy, we present a cloud computing use case where the predictor is deployed. Results indicate that predictions can be accurate within reasonable range, and therefore orchestration algorithms may be improved by accounting for similar prediction models when determining the migration of one or more virtualized functions.

Orchestra: A Customizable Split-Aware NFV Orchestrator for Dynamic Cloud Radio Access Networks

Dynamic Cloud Radio Access Networks (C-RAN) is an emerging wireless architecture that splits radio functions in a hierarchical cloud-based infrastructure. In Dynamic C-RAN, wireless functionalities can be divided into smaller components and distributed along with a hierarchical cloud infrastructure. Network Functions Virtualization (NFV) concepts have been recently investigated to facilitate management-related operations of these wireless functionalities. Despite the advances in providing flexible NFV orchestration for Virtualized Network Functions (VNF)s, there is still a lack of solutions aware of VNFs internal composition, a key characteristic of split virtualized radio functions. In this article, we advance NFV orchestration in a Dynamic C-RAN scenario presenting Orchestra, a split-aware NFVO for Dynamic C-RAN with support to customizable orchestration algorithms for different radio function splits. Orchestra is validated through instantiates and migrates VNFCs following the decisions of personalized orchestration algorithms, with a petty impact in the VNFC's deployment and migration time.

COPA: Experimenter-Level Container Orchestration for Networking Testbeds

As Cloud Computing (CC) branched areas such as Multi-access Edge Computing (MEC) and Fog computing are still on growing research interest. The creation of new tools to improve quality and speed the experimentation in such areas is a general interest. In this article, we propose COPA, an experimenter-level container orchestration tool for networking testbeds. This tool provides a friendly interface for the experimenter test container orchestration algorithms which can start, stop, copy, and even migrate a container from one host to another. COPA also includes network/resources monitoring to feed the experimenter's orchestration algorithm so that it can make decisions based on real-time environment information. Furthermore, the experimenter can automatize the experiment scenario setup and deployment by pre-configuring in COPA. This tool helps the experimenter in testing different scenarios and quickly changing experiment parameters. Considering these features, COPA aims to provide an experimentation architecture to deploy and test container orchestration algorithms. Furthermore, we provide a case study explaining how COPA can be a key tool in the MEC and Network Function Virtualization (NFV) experimentation environments. This tool was already deployed in Federated Union of Telecommunications Research Facilities for an EU-Brazil Open Laboratory (FUTEBOL) testbeds as part of the control framework and was well validated by the project reviewers and partners.

A Service-Defined Approach for Orchestration of Heterogeneous Applications in Cloud/Edge Platforms

Edge Computing is moving resources toward the network borders, thus enabling the deployment of a pool of new applications that benefit from the new distributed infrastructure. However, due to the heterogeneity of such applications, specific orchestration strategies need to be adopted for each deployment request. Each application can potentially require different optimization criteria and may prefer particular reactions upon the occurrence of the same event. This paper presents a Service-Defined approach for orchestrating cloud/edge services in a distributed fashion, where each application can define its own orchestration strategy by means of declarative statements, which are parsed into a Service-Defined Orchestrator (SDO). Moreover, to coordinate the coexistence of a variety of SDOs on the same infrastructure while preserving the resource assignment optimality, we present DRAGON, a Distributed Resource AssiGnment and OrchestratioN algorithm that seeks optimal partitioning of shared resources between different actors. We evaluate the advantages of our novel Service-Defined orchestration approach over some representative edge use cases, as well as measure convergence and performance of DRAGON on a prototype implementation, assessing the benefits compared to conventional orchestration approaches.

Trusted Cloud-Edge Network Resource Management: DRL-Driven Service Function Chain Orchestration for IoT

Private and public networks sharing resources for Internet of Things (IoT) network through network function virtualization (NFV) and software-defined networking (SDN) forms a heterogeneous cloud-edge environment. However, the heterogeneous cloud-edge network faces trust and adaptation issues in resource allocation. To address these two problems, we introduce consortium blockchain and deep reinforcement learning (DRL) to construct the trusted and auto-adjust service function chain (SFC) orchestration architecture. In the architecture, this article integrates the consortium blockchain into the distributed SFC orchestration model to realize trusted resource sharing. In addition, for realizing auto-adjusted service provision, this article designs a dynamic hierarchical SFC orchestration algorithm (DHSOA) based on DRL to minimize the orchestration cost and improve the quality of service. Moreover, considering the dynamics of network entities, this article proposes a time-slotted model to support dynamic service migration which adapts to the high-mobility IoT network. The simulation results show that DHSOA has better performance than the link-state routing algorithm and deep $Q$ -network placement algorithm not only in cost saving of 15.8% and 10.1% but also in time saving of 22.0% and 10.0%.

BOA: batch orchestration algorithm for straggler mitigation of distributed DL training in heterogeneous GPU cluster

Training deep learning model is a time-consuming job since it usually uses a large amount of data. To reduce the training time, most practitioners train the models in a GPU cluster environment in a distributed way. The synchronous stochastic gradient descent, which is one of the widely used distributed training algorithms, has fast convergence rate with the use of multiple GPU workers; but its speed is tied to the slowest worker, i.e., straggler. In a heterogeneous environment, a static straggler, which has not been mainly focused before, has more impact on the performance than randomly occurring straggler. However, most existing studies for straggler mitigation usually consider a homogeneous environment, so their approaches are limited in practice. In this paper, we scrutinize the straggler problem under heterogeneous environment and define static and dynamic straggler from empirical results. Based on this, we propose a novel approach called batch orchestration algorithm (BOA) for straggler mitigation. It adaptively balances the amount of mini-batch data according to the speed of workers. Therefore, BOA can mitigate both static and dynamic straggler in a modern GPU cluster. BOA uses a Min–Max Integer programming to find the optimal mini-batch size, with the hardware-agnostic performance models. For verification, several experiments are conducted on a cluster having up to six GPUs with three types: GTX 1080, GTX 1060 and Quadro M2000. The results show BOA mitigates both types of stragglers and accelerates the training speed with synchronous SGD compared to other straggler mitigation method.

Latency-Aware Resource Orchestration in SDN-Based Packet Over Optical Flexi-Grid Transport Networks

In the upcoming 5G networks and following the emerging software-defined networking/network function virtualization (SDN/NFV) paradigm, demanded services will be composed of a number of virtual network functions that may be spread across the whole transport infrastructure and allocated in distributed data centers (DCs). These services will impose stringent requirements such as bandwidth and end-to-end latency that the transport network will need to fulfill. In this paper, we present an orchestration system devised to select and allocate virtual resources in distributed DCs connected through a multi-layer (packet over flexi-grid optical) network. Three different on-line orchestration algorithms are conceived to accommodate the incoming requests by satisfying computing, bandwidth, and end-to-end latency constraints, setting up multi-layer connections. We addressed end-to-end latency requirements by considering both network (due to propagation delay) and processing delay components. The proposed algorithms have been extensively evaluated and assessed (via a number of figures of merit) through experimental tests carried out in a packet over optical flexi-grid network available in the ADRENALINE testbed with emulated DCs connected to it.

Effectiveness of Segment Routing Technology in Reducing the Bandwidth and Cloud Resources Provisioning Times in Network Function Virtualization Architectures

Network Function Virtualization is a new technology allowing for a elastic cloud and bandwidth resource allocation. The technology requires an orchestrator whose role is the service and resource orchestration. It receives service requests, each one characterized by a Service Function Chain, which is a set of service functions to be executed according to a given order. It implements an algorithm for deciding where both to allocate the cloud and bandwidth resources and to route the SFCs. In a traditional orchestration algorithm, the orchestrator has a detailed knowledge of the cloud and network infrastructures and that can lead to high computational complexity of the SFC Routing and Cloud and Bandwidth resource Allocation (SRCBA) algorithm. In this paper, we propose and evaluate the effectiveness of a scalable orchestration architecture inherited by the one proposed within the European Telecommunications Standards Institute (ETSI) and based on the functional separation of an NFV orchestrator in Resource Orchestrator (RO) and Network Service Orchestrator (NSO). Each cloud domain is equipped with an RO whose task is to provide a simple and abstract representation of the cloud infrastructure. These representations are notified of the NSO that can apply a simplified and less complex SRCBA algorithm. In addition, we show how the segment routing technology can help to simplify the SFC routing by means of an effective addressing of the service functions. The scalable orchestration solution has been investigated and compared to the one of a traditional orchestrator in some network scenarios and varying the number of cloud domains. We have verified that the execution time of the SRCBA algorithm can be drastically reduced without degrading the performance in terms of cloud and bandwidth resource costs.

Availability-driven NFV orchestration

Virtual Network Functions as a Service (VNFaaS) is a promising business whose technical directions consist of providing network functions as a Service instead of delivering standalone network appliances, leveraging a virtualized environment named NFV Infrastructure (NFVI) to provide higher scalability and reduce maintenance costs. Operating the NFVI under stringent availability guarantees is fundamental to ensure the proper functioning of the VNFaaS against software attacks and failures, as well as common physical device failures. Indeed the availability of a VNFaaS relies on the failure rate of its single components, namely the physical servers, the hypervisor, the VNF software, and the communication network.In this paper, we propose a versatile orchestration model able to integrate an elastic VNF protection strategy with the goal to maximize the availability of an NFVI system serving multiple VNF demands. The elasticity derives from (i) the ability to use VNF protection only if needed, or (ii) to pass from dedicated protection scheme to shared VNF protection scheme when needed for a subset of the VNFs, (iii) to integrate traffic split and load-balancing as well as mastership role election in the orchestration decision, (iv) to adjust the placement of VNF masters and slaves based on the availability of the different system and network components involved. We propose a VNF orchestration algorithm based on Variable Neighboring Search, able to integrate both protection schemes in a scalable way and capable to scale, while outperforming standard online policies.

Online Parallelized Service Function Chain Orchestration in Data Center Networks

In recent years, much attention has been focused on deploying service function chains (SFCs), each of which is composed of a set of virtual network functions (VNFs) in a specified order. This is a promising approach for enabling cloud service providers to deploy user service requests more flexibly while saving costs. However, less effort has been directed toward meeting heterogeneous needs, such as high throughput or low latency of user service requests with heterogeneous bandwidth demands, especially in data center networks (DCNs). In this paper, we propose an efficient orchestration algorithm for online SFC requests. It first splits a large flow into a number of subflows and replicates the same number of sub-SFCs. Each subflow is redirected to one of these “parallelized” sub-SFCs, which is termed a sub-user request. Then, each sub-user request is deployed based on a worst-fit strategy, and VNFs in the same SFC are instantiated on the same server to the greatest possible extent. Our algorithm is expected to enable network load balancing, reducing the delay experienced by small flows while improving the acceptance ratio for user requests. Finally, the simulation results show that the proposed algorithm outperforms other comparable algorithms.

Ontology-Assisted Engineering of Cyber–Physical Production Systems in the Field of Process Technology

Future cyber–physical production systems (CPPS) constitute a complex and dynamic network of services and plant components such as actuators and sensors. Consequently, the manual technical specification and design of these systems are a complex and time-consuming task involving extensive expert knowledge. In scope of CPPS, current approaches to reduce the engineering effort focus on manufacturing technology. There are initial approaches in the domain of process engineering available. However, these neither consider the knowledge-supported definition of recipe-based operations nor the assignment of dynamic service networks to process modules. The objective of this contribution is the design of a concept and a systematic approach to automate the engineering of batch process plants respecting dynamic service networks and process modules using a knowledge-based assistance system. For this purpose, a declarative recipe description is combined with an ontological model. This enables an automatic inference of technical requirements. Based on this information, a multistage orchestration algorithm selects and combines process modules and networked services to find appropriate engineering solutions. Finally, a comprehensive case-study demonstrates that the proposed approach is able to automate the target-oriented selection and combination of process modules and service networks.

Semantic adaptation for FMI co-simulation with hierarchical simulators

Model-based design can shorten the development time of complex systems by the use of simulation techniques. However, it can be hard to simulate the system as a whole if it is developed in a concurrent fashion by multiple and specialized teams. Co-simulation, with the support of the Functional Mockup Interface (FMI) Standard, is proposed as a way to promote tool interoperability while protecting the intellectual property of subsystems. The standard allows uniform communication between subsystem simulators, but does not state how the inputs and outputs should be interpreted, nor how the subsystems should interact correctly. Semantic adaptations can be quickly made to correct the interactions with subsystem simulators that were produced with different assumptions, and avoid changing those subsystems, their simulators, or the orchestration algorithm that computes the co-simulation. In this work, we explore how to describe common adaptations and what their meaning is in the context of FMI co-simulation. The result is a sound language that enables the implementation of adaptations with minimal effort. A distinct feature is that it describes adaptations for groups of interconnected subsystem simulators in the same way as for a single simulator, and the implementation is itself a simulator, thanks to a sound definition of hierarchical co-simulation. This work paves the way for research into the correct combination and interfacing of different adaptations.

Cost-Efficient Resource Allocation Method for Heterogeneous Cloud Environments

In this paper we present a novel on-line NFV (Network Function Virtualization) orchestration algorithm for edge computing infrastructure providers that operate in a heterogeneous cloud environment. The goal of our algorithm is to minimize the usage of computing resources which are offered by a public cloud provider (e.g., Amazon Web Services), while fulfilling the required networking related constraints (latency, bandwidth) of the services to be deployed.We propose a reference network architecture which acts as a test environment for the evaluation of our algorithm. During the measurements, we compare our results to the optimal solution provided by an ILP-based solver.

Network service orchestration in heterogeneous 5G networks using an open marketplace

Network service orchestration across heterogeneous networks needs an open marketplace where the services advertised by the providers in different domains can be purchased for short-term or long-term time scales. The authors present two design paradigms and evaluate two corresponding prototypes which provide a framework for network services to be purchased. They compare the two prototypes from the point of view of how effective they are in addressing some of the challenges posed by heterogeneous 5G networks namely programmability, scalability and innovation. They present a network service orchestration algorithm which is advertised as a network service in the marketplace.

SDN-Based Virtual Machine Management for Cloud Data Centers

Software-defined networking (SDN) is an emerging paradigm to logically centralize the network control plane and automate the configuration of individual network elements. At the same time, in cloud data centers (DCs), although network and server resources are collocated and managed by a single administrative entity, disjoint control mechanisms are used for their respective management. In this paper, we propose a unified server-network resource management for such converged information and communication technology (ICT) environments. We present a SDN-based orchestration framework for live virtual machine (VM) management that exploits temporal network information to migrate VMs and minimize the network-wide communication cost of the resulting traffic dynamics. A prototype implementation is presented, and a cloud DC testbed is used to evaluate the impact of diverse orchestration algorithms. Our live VM management has been shown to reduce the network-wide communication cost, especially for the high-cost and congestion-prone core and aggregation layers of the DC. Our results show an increase in network-wide throughput by over six times, as well as over 70% communication cost reduction by migrating less than 50% of the VMs.

Orchestration of Renewable Generation in Low Energy Buildings and Districts Using Energy Storage and Load Shaping

There is increasing penetration of renewable generation in buildings and districts. There are challenges in making the effective use of this generation. The objective of the ORIGIN project (Orchestration of Renewable Integrated Generation In Neighborhoods) is to shape loads so that the fraction of energy consumed that is from local renewable generation is maximized, and energy imported from outside sources is minimized. This paper presents the overall approach taken in the ORIGIN project and explores building physics aspects of solar thermal storage system orchestration. The case study districts are briefly introduced and characteristics of their generation, buildings, districts and shiftable loads described. The orchestration approach taken in ORIGIN is then presented. At the core of the ORIGIN system is the orchestration algorithm which generates informational and control outputs to shape future loads to best meet the objectives. The model based approach used to quantify thermal and electrical load shifting opportunities for pre-charging, coasting or avoiding loads, while meeting thermal comfort and other demands, is described using a solar thermal storage system as an example. The future steps for the ORIGIN project; retrofit of the ORIGIN system into existing districts and potential for other future applications is briefly discussed.