Vendor Column: Advantages of a Scalable Chromatography Data System

Skip to Navigation


  • Published: Jul 10, 2016
  • Author: Chris Stumpf
  • Channels: Laboratory Informatics / Chemometrics & Informatics
thumbnail image: Vendor Column: Advantages of a Scalable Chromatography Data System

In my previous column, I stressed the necessity of comprehensive instrument control and for conversion of legacy data as essential requirements for any chromatography data system (CDS) solution that you decide to standardize. In this column, I will continue the discussion regarding CDS rationalization (evaluating IT infrastructure to see where there are opportunities to standardize onto fewer systems) and focus on scalability. Why scalability? Because if you want to deploy only one CDS solution within your organization then it needs to support multiple users at the same time and grow as your organization grows, whether it is relatively small or global.

First let’s define scalability. From a CDS perspective, scalability is the ability of a CDS system to handle a growing amount of work with the appropriate level of performance as the laboratory grows. A scalable CDS can be deployed in one of three configurations:

(1) A CDS workstation supporting a single user at a time

(2) A small network (“Workgroup”) supporting a laboratory or small department with 5-10 simultaneously logged in users

(3) Enterprise-scale network supporting multiple laboratories or large departments with hundreds or thousands of simultaneously logged-in users across different sites and, in many cases, across geographies

A single CDS workstation is a good way to go if you want to evaluate a new instrument; have a small laboratory; or don’t mind that only one person can only effectively use the workstation at any given time. But if you are planning on future expansion and want to standardize on one CDS platform, then you will need to identify a configurable network product where some of the CDS functionality can run locally (e.g. data acquisition) while others (e.g. data storage) can run on a server. A networked CDS would include either the workgroup or enterprise configurations (2 or 3) highlighted above. Recognize that most typical CDS workstation users choose a workstation based on its functionality and suitability for a particular chromatographic workflow, so it is very important to ensure that any desired workstation functionality is available in the networked CDS. CDS solutions that can keep pace with the growth of a laboratory and scale from a workstation to a small network or enterprise-scale network provide the benefit of required functionality of the workstation and minimizes user training during the transition to a network CDS.

A workgroup or enterprise network CDS configuration can distribute functionality over several different computers. This is known as a distributed computing architecture. For example, the relational database (software application that links 2-dimensional tables of data, consisting of rows & columns, together at run-time) can be installed on one or more servers located in a secure data center, the main CDS application on another server, the instrument interfacing functionality and CDS software clients on yet different computers. Implementing a distributed computing model may sound like a lot of effort, but there are many benefits to a distributed CDS environment - for example workers are more productive when they can process data at a different computer than where they acquired their data and directly access the data from the server. They can work faster when they can query and compare data from a centralized data repository at the same time, negating the need to store multiple copies of raw data. They are more efficient when they can share methods knowing that chromatographic peak integration operations are consistent user-to-user and site-to-site—good if you are working under GxP conditions.

From an IT manager’s perspective, a distributed computing model that leverages contemporary relational database technology (that extends beyond the relational database concept first described by Dr. Edgar F. Codd in a research publication called “System R4 Relational” in the 1970’s) for the CDS translates into improved regulatory compliance, fault tolerance (switching operations to a working server if one breaks), load balancing (maintaining an even computer utilization level so that all users experience the expected level of performance), centralized software maintenance at the server and automated data backups of experimental testing data just to name a few.

Returning to the GxP topic mentioned above, the distributed model for CDS is a good model to support the Electronic Records and Signatures regulation (21 CFR Part 11) for laboratory-based testing systems such as CDS and makes compliance to regulations easier (of course, you still must have sound policies in place in addition to the software’s Part 11 technical controls). A strong theme of 21 CFR Part 11 is trust and consistency. With a network CDS, installation of the CDS software to client computers is centrally managed at the server, ensuring that all clients receive the same “consistent” software package (e.g. software updates) at the same time. After installation, software qualification is performed at each computer to ensure that the software was installed with all of the required files and operates within the established product design. These installation qualification (IQ) and operation qualification (OQ) records are then centrally stored in the relational database at the server for later review. Achieving the same consistency with isolated workstations (as described in 1 above) would be overly burdensome except for very small deployments. The distributed CDS model also ensures data integrity by facilitating physical security. Data stored locally at the laboratory site could fall victim to accidental disruption, deletion or alteration (or sometimes fraud), while in contrast, in a network environment data is automatically captured in a relational database and resides in a data center safely protected from potential hazards (e.g., fires or floods). Hence, network based CDS offers several levels of data security, at the operating system, application and database level, and physical security to ensure that testing data captured is trust-worthy and reliable.

Although there are more benefits to a scalable CDS that I could describe, the functionality and benefits highlighted here give you a flavor for what is possible with a distributed CDS model. So if you are looking to replace an existing CDS that no longer meets your needs or applications; or if you have been asked to rationalize the number of IT solutions that you utilize and need to standardize your CDS operations, selecting a CDS solution that gives you comprehensive instrument control, data conversion for existing/legacy data, and is scalable are worth considering.

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.

Social Links

Share This Links

Bookmark and Share


Suppliers Selection
Societies Selection

Banner Ad

Click here to see
all job opportunities

Most Viewed

Copyright Information

Interested in separation science? Visit our sister site

Copyright © 2018 John Wiley & Sons, Inc. All Rights Reserved