Page 14

Industrial Ethernet Book 103

Technology A CICI Gateway translates and normalizes message payloads as they pass across the domain boundaries of CIP and the Cloud. This crucial operation satisfies a core guiding principle: CIP stays home. Logically, a CICI Gateway can be implemented at the level of a CIP Device, e.g. a CIP Device performs its own CICI Gateway functions. This approach is a two-tier, device direct to cloud connectivity pattern and is not recommended. Cloud In the CICI reference architecture, the Cloud is the Public Cloud. Technically, the public cloud is a complex collection of cost effective, scalable, geographically distributed infrastructure, software and platform services. Conceptually, however, the Cloud should be thought of in simpler terms: as the data collection, processing and analytics platform on which value will be created for the next generation of CIP customers. Value for CIP customers is provided by actionable information which can be used to make asset management and business decisions. Actionable information is derived by the CIP Device related data. Once collected, processed and analyzed, this data results in the ultimate value for ODVA customers: actionable asset information. Cloud-based applications will also consume raw CIP Device data as well as the derived actionable data. Deriving actionable data is the ultimate goal of CICI since this is generally where value is provided to customers. For example, high levels of value to customers due to cost reduction can be achieved in the area of asset maintenance. As device monitoring is increasingly understood and analyzed, maintenance action can evolve from costly run-to-failure (reactive), to more efficient preventive (proactive) to the most cost effective, predictive. Public Cloud Computing This section introduces ‘Cloud Computing’ concepts as related to CICI, so that readers can share a basic understanding of this complex technical landscape. Public Cloud Computing, aka ‘the Cloud’ or ‘Cloud Computing’, represents a significant, disruptive shift from traditional information technology (IT) and software product creation and delivery. In some cases, the Cloud will compete directly with an existing technological model, such as with on-premise hosted Data Center, potentially displacing it. In other cases, the Cloud should be seen, not as a competitive or displacing force to existing technologies, such as CIP, but as a complementary extension and enabler for developing new and different types of applications and solutions which leverage existing technologies. Irrespective of it being a competitive or complementary force, Cloud Computing is disruptive to both technology and business models and brings significant risk for companies and industries attempting to adapt. The simple reason for this is that modeling and implementing Cloud-based solutions such as IIoT is complicated and multi-dimensional. This is not just a question of technology, but extends to business models, the increasing emphasis of OPEX over CAPEX, difficulties with costing imprecision, and the reformation of sales channel and revenue stream, etc. The first big challenge is that the Cloud Computing marketplace evolves daily and presents a bewildering and expanding set of choices among technologies, service offerings, service costing models and vendors. Secondly, the landscape of public Cloud vendors is complex and confusing: small, medium and large vendors compete for business in general purpose and specialized niche service markets. Finally, because of the rapid evolution of the Cloud there is no single path for successful Cloud adoption for a company or even its individual lines-of-business. However, for companies and industries that are able to navigate these complications and can adapt to and leverage the Cloud’s capability new possibilities emerge for developing customer value and driving business opportunity. Cloud Computing offers companies extensive infrastructure, platforms and software services that traditionally have been available only at the corporate, enterprise or data center infrastructural level. These service groupings are referred to as ‘as-a-Service’ resources and are as follows: Infrastructure-as-a-Service (IaaS) is a selfservice model for using storage, servers, networks and related virtual resources. The user of the service is usually responsible for managing applications, data, configurations, licensing and updates of related resources, including in some situations, the operating system. Resources usually have a number of options for performance and quality of service levels and can be very cost effective. A common use of IaaS is for hosting, networking, load balancing and storage for applications. Platform-as-a-Service (PaaS) is vendor managed platforms and middleware on which vendor applications and solutions can be rapidly developed, tested, operated and maintained. Messaging middleware, integration frameworks, databases and business process management are common PaaS services. These services are rapidly leveraged and require no maintenance. Similarly, to IaaS services, PaaS services offer a choice of service level and/or are billed per a consumption-based subscription model. Common uses of PaaS are the deployment of non-monolithic services, integrating decoupled components in a distributed system, storing of quantities of information at Big Data scale and deploying analytics algorithms. Software-as-a-Service (SaaS) is software deployed in the Cloud, yet accessible virtually anywhere by clients. SaaS represents the large and growing field of cloud-based application services and offers as well as new architectural concepts such as microservices. SaaS applications are often accessed via thin-clients such as Web Browsers and mobile applications (native or hybrid) and, therefore, generally require very limited downloaded installation components or none at all. A common use 14 industrial ethernet book 11.2017 SOURCE: ODVA The diagram above shows the Telemetry Information Exchange pattern used in conjunction with the Common Industrial Cloud Interface (CICI) reference architecture.


Industrial Ethernet Book 103
To see the actual publication please follow the link above