Big Analog Data™ Solutions

Publish Date: Mar 14, 2016 | 5 Ratings | 3.80 out of 5 | Print

The Five V's of Big Data

NI is actively working to address Big Analog Data™ end-to-end solutions. But first, consider regular Big Data, which is often defined using a combination of four "Vs":

Volume System is gathering large amounts or volumes of data
Variety Data gathered by the system and analyzed is varied in structure and format
Velocity Data is gathered at a high speed with high sample rates
Value Significant value is derived from the analysis of data; this was previously limited by technology

Another "V" that NI is seeing when working with customers is "visibility." This describes globally dispersed enterprises needing access to the data in multiple locations to both conduct analytics and see results:

 

Visibility Data is accessed or visible from disparate or multiple geographic locations

 

To qualify as Big Data, data does not have to exhibit all five of the characteristics above—just a mix of a few of these characteristics.

 

Where Is Big Data Derived From?

Big Data is derived from a variety of sources that are organized into three major categories:

 

Industry/ IT Data Sources Traditionally associated with Big Data; examples include data derived from enterprise applications, such as ERM, CRM, and HR, and IT data such as events, logs, and inventories.
Social Sources Emerging source of Big Data; more people are generating more social content each day. As users provide candid anecdotal information about products and their experiences with companies online to a broad audience, vendors are looking to take advantage and make sense of this data.
Engineering/Scientific Data Sources Engineers and scientists gather data from natural analog phenomena from the physical world. This is the source of Big Analog Data™ information. That is, it is Big Data derived from the physical analog world that is measured and digitized via analog-to-digital (A/D) conversion.

 

Big Analog Data™ sources are obvious (light, RF signals, vibrations, temperatures, and so on), and they occur naturally in the world or are generated by machinery such as mechanical or electronic equipment. Big Analog Data™ differs in three ways from regular Big Data: It is 1) older (Scientists believe it has been around since the beginning of the universe.), 2) faster as shown by the very high sample and digitizing rates that National Instruments products can collect, and 3) bigger (Sources of analog data are everywhere and are constantly producing data.). 

 

Big Analog Data™ Three-Tier Solution Architecture

NI describes Big Analog Data™ solutions using a three-tier solution architecture, meaning solutions are made up of solution elements that fit into three distinct tiers. These come together to create a single, integrated solution that can be used from the point when data is initially acquired (from a sensor) to ultimately the decision that is made based on this data.

To learn more about how NI products are used in Big Analog Data™ solutions, select the sensors, hardware, or software you are interested in below. 

Click on the NI sensor, hardware, or software featured in the Big Analog Data™ three-tier architecture to learn more about how it contributes to Big Analog Data™ solutions. 

 

Tier 1

The first step in making a decision based on data comes from accurately acquiring it. Analog data comes from acquiring natural phenomena such as temperature, strain, pressure, and so on. Because the varied sources of analog data are so great, each system requires specific sensors and actuators (if used in a controls application) to meet the needs of the type and magnitude of natural phenomena being measured.

Tier 2

This is a valuable piece of an end-to-end Big Data solution because these systems (CompactRIO, PXI, NI CompactDAQ) help connect the benefits of precise data acquired from a sensor to the benefits of in-depth data analysis, usually while the data is real time or early life. The system nodes are network-connected hardware (with associated software) that perform A/D conversion, conditioning, and early analysis of the data acquired and help it move past "the edge" to the switch and server where it is stored and will likely undergo further analysis.

NI Hardware: CompactRIO, PXI, NI CompactDAQ, Wireless Sensor Node 

NI Software: NI LabVIEW, NI VeriStand, NI TestStand

Tier 3

Once through Tier 2, the data moves past "the edge" and usually hits a network switch and then a server where it can be stored and undergo further analysis. This is the point at which customers can realize a great value from the data they’ve acquired and where many engineering, scientific, and business decisions are made. This tier also includes the cloud, a growing and appealing IT infrastructure for NI customers. In addition, it is within the IT infrastructure that much of NI’s systems management and RASM software is run, affecting NI system node (CompactRIO, PXI, NI CompactDAQ) products. This is further explained below.

NI Software: NI DIAdem, NI DataFinder

 

Three Domains of Innovation and Key Considerations When Building Big Analog Data™ Solutions

 

Data Management and Transfer

The value of data is frequently realized beyond the point of acquisition. Thus, the ability to accurately and securely transfer data to other systems for analysis is necessary. The data must be formatted, stored, and easy to sift through, and important features must be made more visible. Engineers need reliable tools for managing, visualizing, and reporting that can quickly access large volumes of scattered data from multiple sources.

Learn about NI DataFinder Server Edition for handling large amounts of data generated during test and simulation.

Data Analytics

Data analysis and visualization are the keys to providing insight into what the data means. The value of data acquired is often found not only after initial analysis has taken place but also after analysis of that analysis. It is of little use to have tons of data if you cannot easily draw conclusions from it to impact your business. This is often when a tactic called "data mining" can be helpful.

Learn more about programmatic analysis using NI LabVIEW.

Learn more about interactive analysis with NI DIAdem.

Systems Management/RASM

Applications often require multiple, distributed system nodes to be used and managed across a network. To ensure these distributed systems are performing as they should, software is needed to configure, deploy, monitor health, inventory, and perform service updates as needed.

  • “RASM” encompasses four separate but related characteristics of a data processing, mechanical, or other physical system (see ni.com/rasm).
    • Reliability: A system operates as intended, without failure or downtime, and satisfies the desired performance requirements.
    • Availability: The measure of how often a system is able to perform its intended function, even in the midst of failures.
    • Serviceability: Features and aspects of the system design contributing to ease of diagnosis and repair.
    • Manageability: The extent to which a system can be controlled, supervised, and monitored.

 

Note that the value of these systems management/RASM tools is many times directly proportional to the number of system nodes (CompactRIO, PXI, NI CompactDAQ) in the solution to be managed. For more information on NI’s leadership in RASM, see ni.com/rasm.

In each of these three areas, NI is investing in innovation to increase the ease and benefit of Big Analog Data™ solutions. Software such as LabVIEW and DIAdem is valuable from the point that data is acquired at the sensor to its early life on acquisition hardware to its archival on a server.

 

Additional Resources

Learn more about Edge Analytics in Big Analog Data™ solutions.

 

 

 

Read a case study from Jaguar Land Rover on how they manage up to 500 GB of data a day collected from 400 engineers. 

 

 

Any media inquires can be directed to beth.williams@ni.com

Back to Top

Bookmark & Share


Ratings

Rate this document

Answered Your Question?
Yes No

Submit