Today’s IoT systems include an increasingly complex mix of new and legacy devices, each with its own unique capabilities and idiosyncrasies. In a perfect world, these intelligent devices would provide complete, consistent interfaces using industry-standard protocols. Unfortunately, this is not the case for many existing devices. Instead, IoT platforms must communicate with each device using its own set of supported interfaces and protocols, even though the data that is returned is sometimes inconsistent, incomplete, or even misidentified. The challenge for the system designer is to derive order from this state of confusion.
Nowhere is this issue of system complexity more evident than in data centers, which host high concentrations of devices representing a multitude of unique equipment types from a wide variety of vendors. We at Emerson Network Power faced exactly that situation as we developed our TrellisTM platform to manage data center infrastructure. Our solution, the creation of a set of “element libraries” containing a protocol map and a data model for each managed device, illustrates the time-consuming but critical task of consolidating and normalizing device data. Over time, we have created models for thousands of unique devices, representing a complete list of the devices we have encountered in the data centers that our systems manage.
Although the task of creating and maintaining consistent device data models can be difficult and resource-intensive, the effort pays off in the ability to gain insight into the operation of complex systems. Using the data gathered through our element libraries, we have been able to create a set of cross-domain management tools that provide significant, quantifiable value to our customers, helping them to lower operating costs, identify system limitations, and improve equipment utilization. These same advantages can be achieved in many other types of IoT systems through applying the concepts of mapping and normalization to communications with and among the end devices.
Steve Blackwell – VP Engineering at Vetrtiv
Steven Blackwell is the Vice President of Engineering for the Data Center Solutions business of Vertiv. He manages a technology and research team focused on data center infrastructure devices and management technologies. His particular areas of interest include device intelligence, system management and control, and remote presence technologies. Steve brings over 30 years of engineering and management experience to the Emerson team, having held design and management positions with Motorola, Adtran, and Centillium prior to joining Avocent in 2003. His development experience includes hardware, firmware, software, and DSP design, and he holds 24 US patents in the fields of data communications and video processing. In previous roles, he participated in data communications standards development, serving in leadership positions in the ITU-T and in FSAN during the development of the foundational DSL standards. Steve is a summa cum laude graduate of the University of Alabama (BSEE, 1986), and he holds an MBA from Nova Southeastern University (1996).
Government, Enterprise, OEM
IoT, Platforms, BlockChain, Data, Devices, Open Standards
CxO, Middle Management, Technical, Business Line Management, Operations
Advanced, Intermediate, Beginner
Retail, Manufacturing, Telecom, Industrials, Healthcare, Consumer, Government / Public Sector, Pharmaceutical / BioTech, Automotive
Join our IoT Community at https://www.linkedin.com/groups/4662022/profile
IoT Grand Slam Virtual Internet of Things Conference