OpenFog联盟边缘（雾）计算白皮书

lead to infrastructure and connectivity limitations and slow adoption and

reduce the value that can be realized via this transformational technology.

Instead, selectively moving computation, communication, control, and

decision making to the network edge where data is being generated is an

emerging area of computer science and electrical engineering. This is called

fog (or edge) computing.

The OpenFog Architecture is a system-level architecture that extends

elements of compute, networking and storage across the cloud through to

the edge of the network. This approach is particularly suited to IoT systems

and accelerates the decision-making velocity. This architecture serves a

interoperability, performance, security, scalability, programmability,

reliability, availability, serviceability, and agility. Proprietary or single vendor

solutions can result in limited supplier diversity, which can have a negative

impact on market adoption, system cost, quality and innovation.

The OpenFog Consortium, formed in November 2015, is based on the

premise that an open architecture is essential for the success of a ubiquitous

fog computing ecosystem for IoT platforms and applications. It is our intent

to ensure the OpenFog architecture results in fully interoperable systems,

supported by a vibrant supplier ecosystem.

The Internet of Things (IoT) is driving business transformation by connecting

everyday objects and devices together and to cloud-hosted services. Current

deployment models emphasize mandatory cloud connectivity, which is not

feasible in many real world situations. Moreover, these connected devices

are creating data at an exponential rate which will drive performance and

network congestion challenges at the edge infrastructure. Current

architectural approaches cannot sustain the projected velocity and volume

requirements of IoT. To sustain IoT momentum, the OpenFog Consortium is

The OpenFog architecture represents a shift from traditional closed systems

and a reliance on cloud focused models, to a new computational model that

moves computation near the edge, and potentially right up to the IoT

sensors and actuators of the network based on workload requirements and

device capability. Called fog nodes, these are not completely fixed to the

believe that OpenFog architecture is complementary to, and an extension of,

the traditional cloud based model as implementations can reside in multiple

layers of a network’s topology which may include a backend cloud.

parameters that can influence the process state. The computational

intelligence can be enabled by a continuum of OpenFog deployments and

backend cloud resources which depend on the domain specific scenario being

realized.

computational layers: regulatory control, supervisory control and decision

support. Conceptually, regulatory control ensures that the process stays

close to the desired state. Supervisory control will ensure that the desired

state is optimized based on the learnings from the current and past states.

Decision support operates on the data accumulated from all the installations

cloud seamlessly. The decision of computation and its associated location

depends on the tolerable latency between a CPP event and actuation.

The OpenFog Consortium invites open participation from across industry,

academia and non-profit organizations which have an interest in the

emerging IoT landscape. The OpenFog Consortium intends to harmonize

with other groups including, but not limited to, the Industrial Internet

Consortium (IIC), ETSI-MEC (Mobile Edge Computing), Open Connectivity

Foundation (OCF) and the OpenNFV.