Cisco Blogs
Share
tweet

Edge-to-Enterprise IoT Analytics – Powered by Cisco and SAS

- February 20, 2017 - 4 Comments

Welcome to Cisco Live Berlin!  Today, I want to introduce the industry’s first comprehensive edge-to-enterprise analytics solution in partnership with SAS.

In my earlier blog, I discussed the promise of connecting the unconnected. Here I will be talking about how to make this a meaningful reality. The Internet of Everything (IoE) will transform every industry, every person’s life in almost every aspect. It brings networking technology to places where it was once unavailable or impractical. The true power of smart, connected devices and the data and insights they generate will create the next era of opportunities.

Cisco estimates 50 billion devices will be connected to the Internet by 2020 and 500 billion devices by 2030. These devices generate data that analytic applications need to collect, aggregate and analyze to deliver informed, actionable insight. The challenge is to build the right digital infrastructure and enable the right set of applications to harness this data. Traditional computing models send the data to the enterprise data center for analysis. This is impractical in many scenarios given the volume of data being produced and the requirement for real-time analysis and response times, often measured in milliseconds. As a result, a new model for analyzing data at the edge of the network has emerged. This model moves the analysis and response close to the devices generating the data minimizing latency and reducing the load on the network and the enterprise data center.

Let me give you a few industry vertical use cases:

Smart Energy:  The Smart Grid, regarded by many as the next generation power grid, uses a two-way flow of electricity and information to create a widely distributed and automated energy delivery network that is safer, more reliable and more efficient, offering substantial benefits for both utilities and customers. Data collected by the devices are used to analyze and understand the energy consumption patterns and adjust the corresponding energy delivery, resulting in greater operating efficiency and environmental goals

 

Connected Automobile: The automotive industry is entering a period of profound change. Autonomous vehicles are close to reality using advanced machine-learning, powered by IoT data. Vehicle-to-vehicle and vehicle-to-infrastructure communications are being developed taking advantage of unified, secure network connectivity enabling transmission of real-time vehicle telematics, GPS tracking and geo-fencing, improving safety, mobility and efficiency, and proactive maintenance which decreases costs and vehicle downtime.

 

Smart Manufacturing: Smart Manufacturing solutions provide intelligent, timely information and collaboration that improves the quality of products and the efficiency with which they’re built. Smart Manufacturing enables continuous improvement of productivity through integration of Six Sigma DMAIC processes (Define, Measure, Analyze, Improve and Control). It also enables more accurate forecasts of product demand, greater visibility into supplier quality level, improved preventative maintenance, better asset management and safer factories.

The Edge-to-Enterprise solution announced today is designed, tested and validated by Cisco and SAS, and has three major tiers: Edge, Transfer and Enterprise, as shown in Figure 1.

  • Edge –  Analytics at the edge means understanding the local context and sending only the important information back to the enterprise. In this case, Cisco 829 Industrial Integrated Services Routers are used. They are designed for deployment in harsh conditions and are running the SAS Event Stream Processing (ESP) client software. The combination enables collecting millions of events per second, filtering the data, analyzing it and detecting patterns of interest in real-time. The output of the analysis defines what alerts to issue, and which data to store and route forward. The Cisco Fog Director Software simplifies the deployment and management of applications and models on the edge routers. Cisco Fog Director is deployed on Cisco UCS C-Series Rack Servers or the Cisco UCS Mini. Advanced data protection and compliance are enforced by Cisco 3000 Series Industrial Security Appliances and Cisco Firepower 4100 Series Firewall appliances.
  • Transfer  – The filtered (or relevant) data is routed from the edge to the enterprise via Apache Kafka: a secure, highly available, distributed message broker. The transfer tier can be physically located at the edge, in the branch office or in the enterprise data center depending on the application use case. Apache Kafka stores the data received from the edge tier and passes it on to the enterprise tier making the system very resilient. Based on the data retention requirements, Cisco UCS C240 Rack Servers or Cisco UCS S3260 Storage Servers are deployed in the transfer tier.
  • Enterprise – Analytics in the enterprise tier enables understanding the global context from multiple edge locations. This tier is powered by the popular Cisco UCS Integrated Infrastructure for Big Data and Analytics platform with Apache Hadoop handling data storage and SAS software stacks including SAS LASR Analytics Server, SAS Visual Analytics and SAS Visual Statistics enabling in-memory processing and visual interfaces for users to create and modify predictive models that meet business requirements. Core to the architecture are Cisco UCS 6300 Series Fabric Interconnects that provide network connectivity, management, and advanced monitoring. The architecture can scale to thousands of servers on demand through the use of Cisco Nexus 9000 Series switches and the Cisco Application Centric Infrastructure (ACI).

Figure 1: Edge-to-Enterprise Analytics Platform

The joint solution is offered as a Cisco Validated Design which provides step by step design guidelines, comprehensively tested and documented, to help ensure faster, more reliable and predictable deployments at lower total cost of ownership. A detailed reference design is depicted in Figure 2.

Figure 2: Edge-to-Enterprise Reference Architecture

Please check out the blog from Oliver Schabenberger, EVP and Chief Technology Officer at SAS for SAS’s perspective of our joint solution.

For More Information
White Paper: Cisco and SAS Edge-to-Enterprise IoT Analytics Platform
Cisco Validated Design: Edge-to-Enterprise IoT Analytics Platform with SAS 
Cisco Validated Design: SAS Visual Analytics with Cisco UCS 
Cisco Big Data Portal 

Cisco Live Berlin Session PSODCT-2020: Scale Big, Scale Fast: Cisco UCS Solutions for Edge-to-Enterprise Analytics. Wednesday, Feb 22, 1:15 p.m. – 2:15 p.m. | Hall 7.1 Breakout Room 714

 

Tags:

In an effort to keep conversations fresh, Cisco Blogs closes comments after 60 days. Please visit the Cisco Blogs hub page for the latest content.

4 Comments

  1. This is great start lets take it to next level

  2. Nice article.

  3. Great work in IoT innovation.

    • +1

Share
tweet