1,720,979 research outputs found
Ensuring Lossless Workload Migration at the Edge with SRv6
No full textIn the context of edge computing, service migration between servers may be needed, for example, to support end user mobility or for load balancing purposes. In this work, we propose a novel solution for service continuity in the presence of workload migration at the edge, which leverages Segment Routing (SR) over IPv6 (SRv6) for packet steering, Software Defined Networking (SDN) and the P4 language for network configuration and programmability, and Network Function Virtualization (NFV) for buffering packets during workload migration. We evaluate our approach on a testbed wherein we use ONOS as SDN controller, and exploit Mininet to emulate the core network of P4 BMv2 switches. We analyze the performance of our SRv6 solution against a non-SRv6 alternative that uses IPv6 routing mechanism. Results highlight the scalability and flexibility of the proposed solution compared to the state of the art
Service Continuity in Edge Computing Through Edge Proxies and HTTP Alternative Services
No full textEdge computing brings computation facilities in close proximity to users, hence paving
the way to a plethora of applications characterized by stringent requirements. Edge systems are highly dynamic, and clients may have to access their edge services at different locations over time. When this happens, it is fundamental to guarantee seamless service continuity, i.e., letting endpoints reach each other transparently and with no or negligible impact on performance. In this work, we propose a service continuity solution for edge environments that leverages an ecosystem of distributed edge proxies as its core element. Edge proxies mediate between client applications and edge services and are dynamically reconfigured by a system orchestrator to ensure service continuity when the proxy for a client needs to change. Our proxies exploit HTTP Alternative Services, an extension of the HTTP standard, to inform clients of the new proxy to reach. Our approach is fully transparent to the application logic and does
not require any non-standard protocol modification. We implemented a Proof-of-Concept and used it to assess our solution over a small-scale testbed. We considered different experimental scenarios and variants of the proposed strategy, comparing it against alternative approaches, namely one where the edge proxy does not change and one based on DNS resolution. Experimental results show the validity and superior performance of the proposed methodology
Server-side QUIC connection migration to support microservice deployment at the edge
Full text linkIn edge computing environments, microservices are typically deployed in the form of containers. To maintain proximity between edge computing services and mobile users, containers need to be migrated between nodes. When migrating containers, however, it is important to consider that they typically have ongoing communications with client endpoints. Moreover, in case of connection-oriented protocols, communicating endpoints share a state (i.e., the connection), which needs to be migrated as well. Connection-oriented protocols like TCP were not designed having connection migration in mind, thus their connections cannot survive a change of IP address or port number. On the other hand, QUIC, a transport protocol recently standardised by IETF, provides a mechanism for client-side connection migration, whereas a server-side connection migration is not yet implemented nor investigated. In this work, we propose an extension of QUIC to support server-side connection migration when a container is migrated between servers. We designed three different strategies, fitting a diverse set of scenarios, wherein the migration procedure is either or not supported by a centralised entity, e.g., an orchestrator. We implemented and verified the proposed extension. Besides, we evaluated it on a real testbed, showing how each of the three strategies is impacted by different container migration techniques and container sizes. To conclude, we compared our solution against two alternatives based on TCP+DNS and MPTCP respectively, demonstrating performance improvements in terms service-migration time
Fog Computing for the Internet of Mobile Things: Issues and Challenges
No full textThe Internet of Things (IoT) conceives a world where everyday objects are able to join the Internet and exchange data as well as process, store, collect them from the surrounding environment, and actively intervene on it. An unprecedented number of services may be envisioned by exploiting the Internet of Things. Fog Computing, which is also known as Edge Computing, was proposed in 2012 as the ideal paradigm to support the resource-constrained IoT devices in data processing and information delivery. Indeed, the Fog, which does not replace the centralized Cloud but cooperates with it, distributes Cloud Computing technologies and principles anywhere along the Cloud-to-Things continuum and particularly at the network edge. The Fog proximity to the IoT devices allows for several advantages that must be continuously guaranteed, also when end devices move from one place to another. In this paper, we aim at examining in depth what it means to provide mobility support in a Fog environment and at investigating what are the main challenges to be faced. Besides, in order to highlight the importance of this topic in everyday life, we provide the reader with three scenarios where there is an integration between the IoT and Fog Computing, and in which mobility support is essential. We finally point out the main research directions in the field
Enabling Application Relocation in ETSI MEC: A Container-Migration Approach
No full textETSI MEC is a standard for edge computing which allows the execution of services - called MEC applications - on hosts in user proximity. One of the emerging concepts within ETSI MEC is that of MEC application relocation, i.e., the migration of a MEC application between edge hosts. ETSI MEC identifies several approaches to relocate a MEC application along with its internal state. However, some of these approaches devote the transfer of the application state to the application itself, whereas others rely on costly virtual-machine migration procedures. To overcome the above limitations, in this work we extend ETSI MEC to support MEC application relocation by exploiting container-migration technologies. We evaluate the performance of our implementation over a small-scale edge testbed, showing the overall benefits of the proposed approach
Virtualization and migration at the network edge: An overview
No full textAs for the Cloud, essential features of Fog Computing are both virtualization and the capability to migrate virtual environments among nodes. In this paper, we thoroughly report both the state-of-The-Art virtualization and migration techniques and their available implementations. In particular, we investigate the aptness of such technologies for a specific layer of the Fog hierarchy, namely the network edge. Indeed, this layer presents some characteristics that distinguish it from the Cloud so that a virtualization or migration technique that represents a good compromise for the Cloud might not be likewise suitable for the edge
Integrating Mobile IoT Devices into the Arrowhead Framework Using Web of Things
No full textInteroperability is one of the main issues concerning the Internet of Things (IoT). The Arrowhead Framework (AHF) is an open-source platform that aims at tackling this problem. In this work, we first integrate a set of legacy IoT devices (i.e., an ECG and a smart plug) into the AHF by leveraging W3C Web of Things (WoT) as an interface between the two worlds. The result is an architecture where AHF consumers are deployed in a local cloud and consume Web Things, which can be deployed on IoT devices as well as on servers at the network edge to provide edge computing services. In this context, IoT device mobility raises a problem. When the IoT device moves, indeed, the edge-hosted service needs to follow it to preserve proximity, which is key to edge computing. In this case, the IP addresses of both the Web Things (i.e., the one hosted on the IoT device and the one hosted at the edge) may change, causing problems of mutual reachability after mobility/migration. WoT does not currently handle this issue. In this work, we extend WoT in this direction and evaluate our approach over a small-scale edge computing testbed
Stateful Function-as-a-Service at the Edge
Full text linkIn FaaS, users invoke remote functions, which encapsulate service(s). These
functions typically need to remotely access a persistent state via external
services: this makes the paradigm less attractive in edge systems, especially
for IoT applications, due to the increased delay and outbound traffic. We
propose to generalize the FaaS paradigm by allowing functions to alternate
between remote-state and local-state phases, depending on internal and external
conditions, and dedicating a container with persistent memory to functions when
in a local-state phase. We present initial results showing that this simple yet
powerful pattern allows to better utilize the available resources, which are
scarce on edge nodes, while significantly reducing tail latencies, which is key
to enable many new applications based on real-time ML, e.g., in smart vehicles
and smart factory scenariosComment: Accepted for publication at IEEE Compute
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
