An Approach to Benchmarking Industrial Big Data Applications

Through the increasing use of interconnected sensors, instrumentation, and smart machines, and the proliferation of social media and other open data, industrial operations and physical systems are generating ever increasing volumes of data of many different types. At the same time, advances in computing, storage, communication, and big data technologies are making it possible to collect, store, process, analyze and visualize enormous volumes of data at scale and at speed. The convergence of Operations Technology (OT) and Information Technology (IT), powered by innovative data analytics, holds the promise of using insights derived from these rich types of data to better manage our systems, resources, environment, health, social infrastructure, and industrial operations. Opportunities to apply innovative analytics abound in many industries (e.g., manufacturing, power distribution, oil and gas exploration and production, telecommunication, healthcare, agriculture, mining) and similarly in government (e.g., homeland security, smart cities, public transportation, accountable care). In developing several such applications over the years, we have come to realize that existing benchmarks for decision support, streaming data, event processing, or distributed processing are not adequate for industrial big data applications. One primary reason being that these benchmarks individually address narrow range of data and analytics processing needs of industrial big data applications. In this paper, we outline an approach we are taking to defining a benchmark that is motivated by typical industrial operations scenarios. We describe the main issues we are considering for the benchmark, including the typical data and processing requirements; representative queries and analytics operations over streaming and stored, structured and unstructured data; and the proposed simulator data architecture.