FAQ – Frequently asked questions
Everything you need to know about Visplore.
Installing Visplore
What are the system requirements for Visplore?
Any Visplore version can be installed on Windows 10/11, including virtual machines. Visplore Professional can also be deployed on Windows Server machines and accessed via remote desktop (RDP). A mouse is not required but highly recommended.
Is Visplore also available for Mac / Linux?
Not yet. Visplore versions for Linux and Mac are planned for the future. For now, Visplore can be installed in a virtual machine running Windows.
Are administrator rights necessary to install Visplore?
- Admin rights are NOT necessary for installing Visplore Free.
- Admin rights are necessary for per-machine installations of Visplore Professional, and all versions that use the Codemeter licensing system. Per-user installations are are possible without admin rights.
What deployment options are there for Visplore?
- Local, e.g. on a Windows laptop or desktop machine, or virtual Windows machine run locally or in local network. (applies to all Visplore versions.)
- Windows server deployment of Visplore Professional, where clients connect via Terminal Server/RDP.
- Windows VMs in the cloud, tested on Azure and AWS (applies to Visplore Free. For Professional, it is supported if the LicenseSpring licensing system is used)
What are the requirements of the data connectors (Python, R, Matlab, OSISoft, ODBC)?
VisplorePy (Python connector):
- 64bit Windows
- 64bit Python 3.5 (or higher) environment, such as Anaconda.
VisploreR (R connector)
- 64bit Windows
- 64bit R, Version 3.5 (or higher) environment, such as RGui, RStudio, etc.
VisploreMatlab (Matlab connector)
- 64bit Windows
- 64bit Matlab Version 2018a or newer
ODBC Connector
- Database must support ODBC, and you must have odbc drivers installed.
OSISoft PI Connector (Visplore Professional only)
- PI Data Archive 3.4.380 and later; PI Data Archive 2012 or later recommended
- PI Web API 2017 R2 (To authenticate with the pi web API the “Kerberos”, “Basic”, or “Anonymous” authentication method should be enabled.)
- PI Indexed Search
- PI Indexed Search Crawler
- Optional: AF Framework
- PI AF Server 2018 (v2.10.0) and later
Which licensing system is used?
Visplore Free requires no licensing system.
Visplore Professional is available with two options:
– cloud-based licenses using LicenseSpring (see https://licensespring.com/). In this case, activation is done by entering a license key.
– using the Codemeter Licensing system by WiBu Systems. See https://www.wibu.com/
Both systems also support license servers for shared (=concurrent) licenses. With Codemeter, these are hosted on-premise by the client, with LicenseSpring they are hosted in the cloud.
Does Visplore require an internet connection?
Visplore Free: no. It checks for new versions at start-up (as many other programs do), but will still start, if it cannot connect to the server.
Visplore Professional: depends on the licensing system used: Versions based on Codemeter don’t need an internet connection. Versions based on LicenseSpring do: for concurrent licenses based on LicenseSpring, every start of Visplore requires internet. There is, however, the possibility to “borrow” a concurrent license for offline use.
Using Visplore
What kind of data can I use for Visplore?
Visplore can be applied to any multivariate data set which is structured as a data table (“data matrix”).
Visplore is specifically optimized for data from time-dependent measurements such as sensor data from machinery, energy, or meteorology. In these cases, rows typically represent observations at specific points in times and columns are the variables per observation. These variables may be quantitative, categorical, or time stamps. More details can be found here.
Visplore Professional additionally supports curve-typed data attributes. A curve may represent a time series per machine operation or an entire spectrum per measurement. Each curve is a set of x/y pairs and can be imported in multiple ways, as described here.
Are there any limits to the data size?
There is no hard limit. Depending on your hardware and type of analysis, a limit for getting reasonably interactive feedback is up to ten million data records (= rows of the data matrix) and up to a few hundred data attributes (= columns of the data matrix). But it really depends on what you are up to do in Visplore.
What if the data is of inconsistent quality?
Visplore can nicely handle missing values, outliers, and other data quality problems. Analyzing such “dirty” data is a major strength of Visplore using features such as interactive filtering, labeling and cleansing. If your analysis requires data to be integrated from multiple sources, however, the integration would require pre-processing outside Visplore.
I have a data source not listed here. What can I do?
For one-time analyses, the most straightforward way is perhaps to simply convert the data to, e.g., a CSV file.
For recurring analyses, Visplore offers APIs for Python, Matlab and R that can be used to read and write data from / to Visplore. For example, it is possible to write a Python application that accesses data from one or more data sources and passes the data on to Visplore. The same application could also read back information such as user-defined data annotations from Visplore and writes it to a data base.
And yes, we will keep adding connectors for an increasing number of data sources in future versions …
How can I share results with colleagues?
You can build your own data story from your interactive visualizations, and annotate the graphs as desired. These stories can be saved as .visplore files (in Visplore Professional), and shared with co-workers, customers, and anybody else. All it takes to open and consume stories, is Visplore Free (yes, this is possible without a paid license!), making a story an ideal deliverable for sharing results within the company, or for consultants to their clients. The benefit over using standard presentation tools is that stories are still fully interactive (zooming in, changing filters etc.). Use stories to make your KPIs reproducible, your analysis steps transparent, and your communication about the data very efficient. In Visplore Professional, stories can also be applied to new data, turning them into ‘live reports’ that can be revisited any time based on the newest data.
And of course, you can export stories as PDF, or export images and processed data to files – also automated via our Python API.
Data Sources
How does Visplore ensure the confidentiality of my data?
Data loaded into Visplore does not leave the computer on which Visplore is running. In particular, Visplore never transfers loaded data to the cloud or other servers by itself. Visplore can also be run completely offline.
How much load does Visplore cause on the data source?
Visplore loads data from a data source (e.g. a file, a database) into memory. The duration and load of this operation depend on the imported amount of data and the performance of the data source. Analysis in Visplore including visualization, selection, cleansing, etc. is based on the representation in the working memory. In particular, analysis does not incur a permanent load on databases.
Is a database required to run Visplore?
No. Databases are one of several possible data sources.
Alternatively, Visplore can load data from files or transfer data directly from a Python, Matlab or R workspace. During analysis, Visplore keeps the data in memory optimized for fast performance.
Can users potentially delete or modify data in our database using Visplore?
No. Databases (including the AVEVA PI system) are accessed on a read-only basis. Processed data can be exported in various ways, but do not overwrite the original data.
