What is your edge? From the cloud to the edge, extending your reach April 2, 2019 Alan.Clark@suse.com
A Definition of Edge “The delivery of computing capabilities to the logical extremes of a network in order to improve the performance, operating cost and reliability of applications and services. By shortening the distance between devices and the cloud resources that serve them, and also reducing network hops, edge computing mitigates the latency and bandwidth constraints of today's Internet, ushering in new classes of applications. In practical terms, this means distributing new resources and software stacks along the path between today's centralized data centers and the increasingly large number of devices in the field, concentrated, in particular, but not exclusively, in close proximity to the last mile network, on both the infrastructure side and the device side.” – State of the Edge 2018 2
Many Edges Core Cloud Edge Data Centers Infrastructure Edge Aggregation Edge Device Edge Gateway Access Edge Device Edge Operating Standalone Edge Core 3
Frictionless Shopping Creating a seamless process from start to finish through the entire customer journey. Modern consumers value convenience and speed, the ability to easily find products and a fast checkout CC BY 2.0 Guiseppe Milo,
Augmented Workforce “ One of the new rules for the digital age is to expand our vision of the workforce; think about jobs in the context of tasks that can be automated … and the new role of human skills…” Public Domain
Industry:Oil pipeline Shannonpatrick17 CC BY 2.0
Manufacturing Collecting data from manufacturing machines in real time, enabling the company to accelerate failure responses, reduce unplanned downtime, improve factory- floor maintenance, and increase production efficiency Steve Jurvetson, CC BY 2.0
More Case for Edge Supply chain Health Care Smart cities • Telemedicine transportation solutions Remote security • Driverless cars and trains Video analytics Mobile Connectivity Location services Remote Location (Finance/Retail) Optimized local content Utilities (water, gas, electric, distribution and data caching facilities management) Energy Drones
Common Thread For The Edge Cases Distributed infrastructure demands a different architecture Characteristics • Massive data growth • Workloads move closer to the end user • Software applications tap into local content and real-time information Sense Process Act - Meet application latency (< 20 ms) requirements • Deliver an improved end user experience • Do not place unreasonable demands on connectivity to the core - Network needs to be smarter - Work in conjunction with the core capacity
Common Thread For The Edge Cases Distributed infrastructure demands a different architecture Demands Characteristics • Massive data growth • High service uptime • Workloads move closer to the end user • Streamlined operation - Admin/Operator - aggregate, supervise and • Software applications tap into local content expose massively distributed infrastructure and real-time information resources - Meet application latency (< 20 ms) requirements • DevOps • Deliver an improved end user experience - Implement new kinds of services that may be • Do not place unreasonable demands on deployed/managed on-demand connectivity to the core • Critical infrastructure - Network needs to be smarter - Enable high performance, reduce latency, improve - Work in conjunction with the core capacity availability, scalability, address security need, improve fault management
… More to consider… Vast numbers of deployments Consistency with the core People time and skills demands • Think Control - Unified orchestration Remote or unreachable • operational mechanisms environments • Service management • Zero touch life cycle management • Physical designs – limited power, processing - Easy deployment capacity, security, dirt, humidity, vibration - Low touch manageability Moving data isn’t free - Maintainability • The data transfer & compute requirements - Rapid response to events vary by network layer - Fast recovery - Fault management
Open Source Edge - Chaos? Edge Computing Group IEEE-SA(1934 ) ETSI-MEC
Open Source Edge Edge Computing Group IEEE-SA(1934 ) ETSI-MEC
Open Source Edge Creating an open architecture for fog Edge Computing Group computing that is necessary to enable interoperability and IEEE-SA(1934 ) scalability. ETSI-MEC
Fog computing is a: A horizontal, system-level architecture that distributes computing, storage, control and networking functions closer to the users along a cloud-to-thing continuum. An extension of the traditional cloud-based computing model • Where implementations of the architecture can reside in multiple layers of a network’s topology Preserved benefits of cloud • Including containerization, virtualization, orchestration, manageability, and efficiency Architecture Pillars of OpenFog • Security, scalability, openness, autonomy, RAS (reliability, availability and serviceability), agility, hierarchy, and programmability.
Fog Deployment Models Intelligence Creation Raw data & data processing
Open Source Edge Encompasses various approaches to disperse Information Technology (IT), Communication Technology (CT) and Operational Technology (OT) Services Edge Computing Group through information messaging infrastructure as well as legacy and IEEE-SA(1934 ) emerging multi-access networking technologies ETSI-MEC
Open Source Edge Creating a standardized, open environment which will allow the efficient and seamless integration of applications from Edge Computing Group vendors, service providers, and third- parties across multi- IEEE-SA(1934 ) vendor Multi-access Edge Computing platforms ETSI-MEC
ETSI MEC Specifications Best Practice and Guidelines Radio Network Information API Use Cases and Requirements Mobile Edge Platform Application Enablement Technical Requirements Mobile Edge Management; Reference Architecture Application lifecycle, rules UE application interface and requirements management Bandwidth Management API System, host and platform UE Identity API management Location API General principles for MEC General principles for Mobile Service APIs Edge Service APIs
Open Source Edge …standardizing on how REST APIs are described. Edge Computing Group IEEE-SA(1934 ) ETSI-MEC
Open Source Edge Edge Computing Group IEEE-SA(1934 ) ETSI-MEC
LF Edge Umbrella Akraino: • An open source software stack that supports high-availability cloud services optimized for edge computing systems and applications. The Akraino Edge Stack is designed to improve the state of edge cloud infrastructure for enterprise edge, OTT edge, and carrier edge networks. Edge X Foundry: • A loosely-coupled microservices framework with a focus on the IoT Edge. It simplifies the process to design, develop and deploy solutions across industrial, enterprise, and consumer applications. Edge Virtualization Engine: • A new project to develop new levels of control of cloud native applications through virtualization of on-prem edge devices. Home Edge: • Intelligent home edge computing open source framework, platform and ecosystem running on a variety of devices with interoperable, flexible, and scalable edge computing services with a set of APIs that can also run with libraries and runtimes.
Open Source Edge Edge Computing Group IEEE-SA(1934 ) ETSI-MEC
Eclipse IoT 24
Open Source Edge Edge Computing Group IEEE-SA(1934 ) ETSI-MEC
StarlingX
SUSE Software-defined Infrastructure and Application Delivery Approach Application Delivery Container Management Platform as a Service SUSE CaaS Platform SUSE Cloud Application Platform Software-Defined Infrastructure Private Cloud / IaaS Infrastructure SUSE OpenStack Cloud & Lifecycle Management Compute Storage Networking SUSE Manager Virtual Machine & SUSE Enterprise SDN and NFV Public Cloud Container Storage SUSE Cloud Service Provider Multimodal Operating System Program SUSE Linux Enterprise Server Physical Infrastructure: Multi-platform Servers, Switches, Storage
Leverage SDI To The Edge To meet the demands: Massive Data Move workloads Apps tap User experience Connectivity 28
References [1] Cloud Edge Computing, Beyond the Data Center, https://www.openstack.org/assets/edge/OpenStack-EdgeWhitepaper-v3-online.pdf [2] http://archive.sltrib.com/article.php?id=1383502&itype=cmsid [3] https://nrf.com/blog/new-kind-customer-whats-driving-retail-evolution [4] https://www2.deloitte.com/insights/us/en/focus/human-capital-trends/2017/future-workforce-changing-nature-of-work.html [5] https://medium.com/cerebrus/proximity-computing-edgy-edgier-edgiest-94cc92971cf6 [6] ETSI MEC Reference Architecture: https://www.etsi.org/deliver/etsi_gs/MEC/001_099/003/01.01.01_60/gs_MEC003v010101p.pdf [7] https://www.ibm.com/downloads/cas/NDE0G4LA [8] https://www.suse.com/media/success-story/knorr_bremse_ag_ss.pdf [10] https://www.etsi.org/deliver/etsi_gs/MEC/001_099/003/01.01.01_60/gs_MEC003v010101p.pdf
Recommend
More recommend
