Reliability HotWire

Reliability HotWire

Issue 132, February 2012

Hot Topics

Challenges in Selecting an Integrated Enterprise-wide Reliability Solution: An IT Perspective

Guest Submission

Jassem Al-Salem
Senior System Analyst, Kuwait Oil Company

Over the past few years, reliability methods have become more and more crucial and essential tools in the oil and gas industry. In the Middle East region specifically, their importance is becoming more evident, and more and more users are asking for the tools in order to utilize these methods. From an IT perspective, our purpose is to support the users and assist them in selecting the right tools, which would also meet our requirements. This article describes the challenges faced when selecting an integrated enterprise-wide reliability system, based on my experience at Kuwait Oil Company (KOC), as part of the team assigned to make this selection.

Introduction

In our quest for a reliability solution at KOC, we run into the common question these days, which goes beyond reliability-oriented systems: Is it the right decision to go for an enterprise-wide solution or for standalone software?

Quite often, there are conflicting requirements and needs, from both engineering and information technologies (IT) standpoints, which necessitate a compromise between functionality, usability, ease of initial deployment, long-term support, cost, etc.

The information system team always has a different perspective, mainly on the software and hardware deployment, networking, licensing and security. Specifically, the main points under consideration from an IT standpoint are:

  • Data protection
  • Reduced maintenance cost
  • Rapid deployment
  • Scalability
  • Integration with real-time information
  • Virtual server environment and storage
  • Knowledge sharing across the company
  • Storage management
  • Enterprise-wide data integration
  • Backup and recovery
  • System upgrades

On the other hand, the users’ needs also need to be fulfilled, and we must ensure that the chosen system provides the necessary capabilities and analytical power to deploy the required reliability methods. So the task is to identify the users’ basic requirements and goals and then come up with a solution to satisfy both IT and users. In our case, and from a top level perspective, the following user requirements were identified:

  • Ease of access
  • Analytical power
  • Seamless analyses
  • Seamless reporting
  • Different levels of reporting
  • Knowledge management
  • Integration with real-time information

There are obviously many details on the above mentioned requirements, but they are beyond the scope of this article.

Once the IT and user requirements were determined, the common objectives were identified, as shown in the figure below.

Venn Diagram

Based on the common requirements it became apparent that some form of an enterprise solution would be needed. Before making a final decision, though, additional considerations must be taken into account.

Hardware Considerations

In choosing the right system, the impact on hardware is very critical. From a hardware perspective, deciding on an enterprise solution over standalone or client-based will result in additional overhead. Specifically, the following items need to be considered as part of an enterprise system:

  1. The system should provide high availability and be scalable and secure so that it is accessible to any company users and its business partners to carry out their business functions based on their access profile, role and authorization level.
  2. Data volumes, database size and potential growth, the expected load, transaction rates and the like.
  3. Database instances (environments) to be maintained (such as configuration/unit test, development/unit test, parallel production test, training, production, and the like.)
  4. Components required to implement secure environment and communications with business partners including, but not limited to, digital certificates, SSL accelerator cards, and the like if applicable.
  5. Additional licenses for enterprise databases.

In many cases, a standalone or client-server approach will provide better performance with less overhead. In this case, a balanced approach was taken by enabling the provision of software installation on client PCs with a network-wide license or terminal server approach to address the IT department's desire for a scalable, manageable solution while still satisfying the users’ demands for functionality, usability and performance.

Web-based?

Web-based systems offer ease of deployment and access. In many ways they offer the ultimate enterprise solution. However, as of today many of the capabilities currently offered in desktop/client-server applications cannot seamlessly be replicated in a web environment.

Conclusion

The IT department has a lot to say on selecting a centrally deployed system, which is really valuable for the long-term maintainability perspective.

The company is going through frequent server and desktop upgrades as it looks for technology that can reduce the complexity and labor required to complete the task successfully and securely. IT is also concerned with handling large customer volume and large infrastructure. Deploying enterprise-wide infrastructure will enable the company to lower the administrative costs, improve the scalability and achieve better reliability, and to increase security by integration with Windows Active Directory and mail server.

Considering all of the requirements and demands, the company adopted a balanced approach in the architecture, which is a "hybrid" solution comprising:

  • A web-based enterprise system for the data integration and reporting of key performance indicators (KPIs) for asset performance management, which provides ease of access, supports individual user access levels and is "light" on hardware requirements and easy to update.
  • Client-server applications for advanced reliability analysis, which integrate with the web-based system. A terminal server approach was selected to address the IT department's desire for a scalable, manageable solution while still satisfying the users’ demands for functionality, usability and performance.
  • High performance virtual servers were selected for asset performance management and client-server architecture reliability software applications for easy migration and availability.
  • Servers are connected to storage via fiber channel-based storage network (SAN architecture)

Last but not least, for a successful implementation of a maintenance and reliability solution, it is necessary to involve teams from the user group, IT, the group responsible for the maintenance management system and management staff in the early stages of the project — during the basic design, bidding, evaluation and implementation. All the teams also need to provide a familiarization program and training during the project life cycle to get more quality participation.

In general, the enterprise-wide system is likely to be deployed if dealing with:

  • Process-oriented analyses that require input and review by multiple people.
  • Cases where the organization will benefit from centralized data storage.
  • Multiple users who need to log in to the system from various locations and query or update the shared information.
  • A need to provide consistency, corrective actions, "knowledge base" of known issues, etc.
  • A need for rapid restores.
  • A need for single point remote maintenance.
  • Any situation requiring low maintenance and operating costs. Enterprise-wide systems provide this mainly due to having little or no software that needs to be installed or updated on each user's computer. This characteristic is understandably attractive to many IT departments!

About the Author

Jassem Al-Salem has worked as a senior system analyst for Kuwait Oil Company for the last 30 years, serving in IT, handling help desk, engineering software, support for real-time systems and enterprise system deployment.

He can be reached at jsalem@kockw.com.