The cloud has multiple tenants, meaning that one architecture hosts multiple customers’ applications and data. The “noisy neighbour” effect occurs when an application or virtual machine overuses the available resources, causing network performance issues for others on the shared infrastructure. When one application or instance uses too much data or bandwidth, other applications suffer from slow speeds or latency. This problem can affect web hosting, databases, networks, storage and servers.
Last semester, the IBM Research Lab in Haifa cooperated with the Technion Israel Institute of Technology to develop a project in which the students investigated the problem of detecting noisy neighbours in a network using Skydive analyzer. (more…)
5G IA is hosting the 7th Global 5G Event– ‘Creating the Smart Digital Future’ on 17-18 June in Valencia, Spain in co-operation with the leading 5G visionary organizations from Brazil, China, Japan, the Americas and the Republic of Korea. Details available online at5g-ppp.eu
From 2nd to 4th April 2019 the SliceNet project meeting is taking place at Ericsson Italy in Pagani, between Mount Vesuvius and the City of Salerno. On 3rd April, Alessandro Pane, Head of Ericsson R&D Italy, demonstrated his great interest in SliceNet by participating in the meeting. In his presentation he talked about the relevance of SliceNet in the context of Ericsson Italy’s overall R&D activities related to 5G and network management.
On the last February 5 2019, Salvatore Spadaro from UPC presented the paper entitled “Supporting QoE/QoS-aware end-to-end network slicing in future 5G-enabled optical networks” at the Photonic West 2019 conference, held in San Francisco (US) on February 2-7, 2019. In particular, the paper presents an architecture design enabling network slice provisioning for 5G service chaining in multi-segment/multi-domain optical network scenarios. The design is enriched with an actuation framework able to maintain the desired QoS for the provisioned end-to-end network slice.
IEEE Transactions on Broadcasting (TBC) has recently accepted a SliceNet contribution “Enable Advanced QoS-Aware Network Slicing in 5G Networks for Slice-Based Media Use Cases”. TBC is an internationally leading journal on broadcasting, and communications and networking in a wider context. The paper presents some of the latest achievements from SliceNet in the design, prototyping and empirical evaluation of advanced network slicing with controlled Quality of Service (QoS) in 5G networks for media use cases.
In particular, a set of innovative SliceNet enablers are described in details, covering network slicing over radio access network and core network, a low-latency multi-access edge computing (MEC) platform, a hardware-accelerated programmable data plane, plug and play (P&P) control for customised network slicing, and a one-stop interface for vertical businesses. The paper combines all these key enablers in the SliceNet framework to address the mission-critical application requirements in the SliceNet media-rich eHealth use case.
5G PPP PHASE 2 KEY ACHIEVEMENTS, prepared by the 5G PPP Technology Board, are availablehere
SliceNet is listed in the achievementshereunder the 5G Multi-Domains Multi-Tenants Plug & Play Control Plane and Slicing Control icon, where the project results are described under Multi-Domain Multi-Tenants Network Slicing and Plug & Play Control for Verticals. SliceNet defines business roles and enabling mechanisms to achieve cross-domain network slicing for the agreed Service Level Agreement. Furthermore, SliceNet Plug & Play control plane provides a per-slice control environment, which offers verticals and consumers, a significantly enhanced degree of flexibility for deploying services.
SliceNet has recently proceeded with the design and implementation of a 5G RAN-Core slicing-friendly infrastructure supporting control and data plane programmability based on the FlexRAN and LL-MEC Mosaic5G platforms on top of the OAI RAN and OAI CN deployment. Ongoing work includes the design and prototype implementation of a RAN monitoring framework extension over FlexRAN that will provide raw material to QoE monitoring and machine learning-based models for resource optimization.
There are two network cognition strands: first, an integrated methodology approach to cognitive network & slice management for designing, implementing and running cognitive models; second, on the aspects of cognitive control to support QoS/QoE in RAN slicing.