
6 Steps to a Successful FlexiCapture Distributed Install
Read our White Papers and other technical articles to learn about OCR, data capture, and RPA technology.
1. Understand the Interaction of the System Components
FlexiCapture distributed has several primary components that will typically run on separate servers. Understanding the interaction of the components is critical to understanding and fully utilizing the FlexiCapture software.
The main FlexiCapture Application Server contains the system monitoring components along with the websites and user security. Also installed on the Application Server, will be the Processing Server which is responsible for the distribution of workloads off to the Processing Station(s). Processing Stations are used for their CPU resources to perform the actual OCR recognition work and are intended to be separate environments from the Application Server. FlexiCapture Distributed also requires the use of a SQL Server Database. This database can be collocated on the main Application Server, however it is strongly recommended that the database is hosted on a dedicated, enterprise grade database server to ensure that proper performance and database maintenance policies are in place.
2. Follow the Hardware and Prerequisite Recommendations
FlexiCapture is a powerful enterprise level software suite. It requires powerful hardware that will allow the software to run to its fullest potential. Under-sizing the hardware for the volume of data will result in instability, errors, and a lack of throughput. A common mistake is to allocate 100% of a server’s resources to Document Recognition; this will starve the system for overhead resources, which will often result in errors or general instability. It is strongly recommended to error on the side of caution and size the server for future volume up front, rather than addressing a lack of resources retroactively.
The ease of procurement of a simple virtual server can be a tempting starting point for a new implementation. However, after the system is in production, making significant hardware changes can be an arduous process through outage planning, procurement, and change control policies. While virtual environments are certainly supported, it is up to the customer to ensure that their virtual servers provide the same processing power and performance as the equivalent physical server in order to ensure proper throughput.
3. Read and Understand the Documentation Prior to Installation
ABBYY FlexiCapture has excellent documentation. The main System Administrator’s Guide deals primarily with the installation of FlexiCapture server components, and details the Windows Server prerequisites that are mandatory for a successful installation. It is absolutely critical that the prerequisites are followed exactly as detailed in the Admin guide, as any missing components will result in problems during the installation.
Consideration must also be given to the prerequisites for any of the Workstation software that will be installed on the Application Server. For example, the Project Setup Station application requires Microsoft .NET 3.5 to be installed. This is often missed during the server installation and will result in an error when creating Document Definitions.
Within each of the FlexiCapture software components are excellent help files which contain the solution to most common configuration questions. It is recommended that FlexiCapture users and admins alike familiarize themselves with the FlexiCapture Help files.
4. Establish Database Credentials Prior to the Installation
During the FlexiCapture Server installation, the final step is to create the database and establish the database connection. The recommendation and preferred method is to use SQL Server authentication and an SQL user with sufficient rights to create a new database. If desired, once the FlexiCapture database is created and the connection established, the SQL permissions can be reduced to satisfy internal security policies.
5. Locate and Utilize Master Data (for Invoice Recognition only)
Successful invoice recognition requires the use of ERP system Master Data. Prior to the installation and configuration of FlexiCapture, it is recommended that the current Master Data is known, acquired, and accessible. Once the FlexiCapture installation is complete, the Master Data can be linked to the FlexiCapture project where the Vendor, PO header, PO Line Items, GL Codes, and Cost Centers data can be used in order to ensure reliable Invoice recognition. The integrator assisting with the FlexiCapture configuration will be able to advise the desired format and location for the data, and how to best utilize it for recognition.
6. Determine the Export Location for the Extracted Data
FlexiCapture can export the extracted data to database tables and images can be exported to the file system or database tables as BLOBs. Data can also be exported to just about any 3rd party application such as SharePoint, Sage, or QuickBooks. Prior to the installation and configuration of the FlexiCapture software, it is recommended that any external system credentials, UNC share paths, URLs, DB connection details, or other system 3rd party specific details should be determined and documented in the project notes.
Use of external resources often requires the Processing Station to run as a named user with permissions to the desired resources. Unless it was changed during the installation, the software defaults to the Window’s user “Network Service”. In most cases, this user will not have permissions to resources outside the localhost.
ABBYY FlexiCapture is a complex and powerful software that requires a commitment to gaining knowledge and understanding in order to fully utilize it. Taking a proactive approach to preparing for the installation and following the suggestions in this posting will ensure a quick and painless ABBYY FlexiCapture installation and setup.