The Path of the Historian
In Part One: What is a Data or Operational Historian, it was shared how Historians evolved from chart recorders; developed in order to view historical trends in data. Using data historians, we are able to track, record, and store different gauges and readings for machinery including but not limited to; temperature, units produced, pressure levels, flow rates, energy consumption, and more. This helps optimize processes and use of machinery and also can be used to ensure compliance to company or regulatory policies. We also discovered though, that while being a great tool for users – it did leave something to be desired.
A Future in Need
Upon looking at the strengths and weaknesses of the given options from Part Two: How To Use Historian Data and Existing Options, one can surmise that in order to create the perfect device for users, it would have to optimize the following:
- Read/Write Speed
The speed of data input and processing at which Historians work is undeniable. In comparison to just about every other type of traditional software/option, it is able to fulfill hundreds of thousands entries every second. With this kind of speed, data historians can find enterprise-wide use. With multiple sensors, pulling from a variety of check-points on every machine on the plant floor, all desired statistics could be gathered. Efficiency of data storage for historians also ranks highly in comparison to the most common alternative; relational databases..
- Analytics and Processing
Possibly the greatest missing factor of the Historian is the ability to create comparative relational analysis in and of itself. This requires a third party tool and generally involves the grueling task of transferring large volumes of data. In this situation; we took a look at relational/traditional columnar database systems in which although populated significantly less data, were made for this type of data processing. Relational databases offer a very functional structure of data that is easy to understand, and easy to use and very familiar among users but take significantly more time both storing and retrieving data.
- Cost Effectiveness
In the current state of price availability based on user licensing, amounts of tags, and storage, while still having the limitations and gaps previously discussed; the data historian is not always the best option available. Especially when keeping in mind the dynamic use of different database software or opening up the possibility of using something open-source. Having a combination of the previous items is not always realistic either; when thinking about the possible complexity of having to deal with the convolution of using different tools.
Is it even possible? To optimize read/write speed, analytics and processing, while still being able to be cost effective in a single tool sounds unattainable. Each option is a viable choice for a reason, right? This is generally the case until the further advancement of technology which breaks or surpasses previously held limitations…
The Development of Time Series Database Management and IIoT
With great strides being made in technology; the IoT (Internet of Things) and IIoT (Industrial Internet of Things) have made ebbs and flows in many different industries, causing many different businesses to adapt if they wish to stay afloat; Data being one of the industries. Machbase has proven itself to be an innovative leader in the pack with the Database Management System optimized for IoT and Big Data compatibility.
Data Entry Up to Millions of Inputs Per Second
With great strides being made in technology; new time series databases are becoming faster than their counterparts; Machbase having the ability to track and record more than a million entries each second. The architecture of it is optimized for sensor data processing. In addition to the ability to track data, it is also able to be compressed in real-time to store more efficiently, and also to process queries at faster speeds as well..
- Real Time Data Analytics; One Stop Shop
Along with data entry speed; newer Time Series Database Management systems are able to do real time data analytics. There is no longer a need to have different tools for different functions. Especially with software being able to integrate with IoT data devices; it gives the ability to more easily collect additional data from different devices for analytics. In a world with increasingly cheaper and cost-effective sensors, the price model for data historian licensing becomes more obsolete. Data written/pulled is not just faster than before, but more current for real time decision making.
- Reliability & Support
With the reliability & support of proprietary database management systems, the technology is advancing and becoming more scalable. With their cluster-supported HA (High Availability) technology, Machbase is able to engage in continued service, even in the event of server failures. Adding additional servers not only increases linear data performance, but creates added reliability.
Will Historians Weather the Storm?
Data Historian software has played a big role in the development of Big Data. Even up to now, Historian data is stable, easy to implement for companies and easy to maintain. Historians still rank highly in high speed data input in comparison to most other available options. Also, with data historians having been utilized for so long, added with the idea that companies are comfortable using them and change is sometimes resisted; they most likely won’t go away as an option for manufacturing plants. It has already come to a point though, that Historians may not be the optimal choice in the long run without adjustments and further development in order to compete.
One could point to Traditional chart recorders which are still finding their use for some companies. If chart recorders have survived and are still around after all the technological advancement of the past 3 decades, Historians could probably be able to keep a place in the industry of machine manufacturing.
The future though, seems to require more adaptation and a compatible vision in conjunction with the IIoT.