Home arrow Master Plan arrow Master Plan Phase I arrow Evaluating OSS Solution Areas
Evaluate OSS distributed enterprise Print

Enterprise networking is becoming more critical to an organization’s environment than ever before. They must be able to manage and consistently troubleshoot remote as well as local segments of widespread company networks to continue business. To top it off, the amount of data flowing across networks is growing rapidly. Organizations deploying central server models are starting to experience bottlenecks caused by these networks. One solution is to incur larger network charges to increase network bandwidth in order to eliminate the bottlenecks. Alternatively, they could implement distributed enterprise solutions to reduce the amount of data that need to flow over the already congested networks.

Distributed Enterprise solutions are typically beneficial for large agencies that conduct its business in multiple remote locations. Due to the expanse of operations across the country, network costs can easily be very high and some form of control is necessary. Distributed Enterprise solutions seek to have computational transaction processing capability both at the central site as well as the remote site. This way, the amount of data flows can be reduced as each site can perform its own processing. An index of the results or the results itself is sent back to a central store for access by other locations or users. They typically have highly replicated servers all performing the same function in various distributed locations (outlets, branches, etc.) within a large enterprise.

OSS Distributed Enterprise Benefits

Decentralization of IT resources

By decentralizing data and compute resources closer to the clients, organizations will be able to increase end user application response times. This in turn would promote user productivity within the organization. It would also allow for more complex functions to be developed and made available to users as the solution of the network bottleneck is addressed.

Empowerment of decision making to remote locations

In the past, only users with access to business tools and data were informed enough to make decisions. Since these solutions were normally house in a central location (e.g. headquarters), decision making involved various processes between the remote and central sites. Once the users in remote sites have access to the necessary tools with reasonable response times provided by distributed computing solutions, they will be better informed to make decisions. This will also increase the organizations productivity as long approval processes can be reduced.

Increase transaction processing

With centralized solutions, organizations with remote branches would have to have some process to gather transactional data (e.g. payments, transaction records, visa/passport/IC applications, etc.) that would have been collected by the remote branches. Once gathered, the information would have to be verified and entered into the central data store. Since organizations cannot frequently perform these procedures, the central data store would only be updated at intervals or batches. Users analysing the data would only see what was available since the last update. With distributed enterprise solutions, transactions can be performed remotely and the final record updated to the central database in real time. This gives analysts a better view of the current data gathered.

OSS Distributed Enterprise Components

End-User Device Tier

This is the lowest of three tiers within the distributed computing solution area. This tier would consist of devices and computers that end users would use to interact with the application. Remote site client machines, end-user devices (e.g. tablet PCs, touch screens, self-checkout devices, etc.), automated teller terminals, staff workstations and handheld devices (e.g. PDAs, pagers, mobile phones, etc.) would fall under this category.

There are a range of operating systems available (Windows, Linux, PalmOS, UNIX, etc.) for the devices in this tier but the choice of which would depend on which operating system the proposed application’s client program will be able to run. Certain devices, particularly the hand-held ones, due to their specialized embedded microchips, would have a very limited choice of operating systems that would run on them./

However, if the solution is based on open standards, typically web-based, then the dependencies on the client operating system can be reduced as long as it would support a capable browser. In this situation, agencies can deploy OSS operating systems (e.g. Linux, FreeBSD, etc.) using OSS web browsers (e.g. Mozilla).

Distributed Server Tier

This middle tier in the three tier model consists of IT infrastructures deployed at the remote locations. These servers would provide site infrastructure functions and be replicated in each location. There may be a need to use open standards for easier integration with other systems in the other tiers.

Having servers in distributed locations can sometimes cause server management issues for the agency. OSS tools like OpenSSH, Webmin and Nagios are available to allow system administrators to remotely monitor and manage these servers.

Centralized Server Tier

This final tier would consist of the centralized servers housing the central data stores of the application to support corporate-level data requirements. Management of the deployed application (e.g. user management, access rights, functionality enablement, etc.) could also be done with these servers. Legacy systems like existing mainframes that the distributed enterprise solution integrates with would also fall into this tier.

Agencies should consider developing applications in such a way that they would have a broader range of upgrade options in the future. For example, should an agency choose to develop an application on Linux, that agency would have the benefit of running that application on the large range of hardware platforms that Linux would support. This gives them the flexibility to determine the most appropriate platform to suite their needs and budget.

OSS Distributed Enterprise Success Factors

Existing environment assessment

In order to reduce resource issues, the agency’s existing environment should be studied closely. These studies help in collecting all the required data for designing a distributed enterprise solution like server hardware details, network utilization, operating systems and running applications. This data will help solution architects design and size an adequate solution which will sustain the growing need of organization.

Technical Feasibility Check

Before finalizing on the solution, a technical feasibility check should be performed. This should involve the existing system administrators and engineers designing the solution. The check would highlight any interoperability issues and would ensure a smooth deployment.

User Awareness

User awareness on solution should be built up with constant knowledge transfer sessions with the users and system administrators throughout the implementation. This ensures that users will be appreciate the benefits of the system and be ready to use it when it is fully implemented.

Define Data Flows and Formats

The flow of data between each component of the proposed distributed enterprise solution must be clearly defined in order to ensure that proper standards are followed when communication occurs. Additionally, data format requirements should be specifies so that necessary conversions can be made at appropriate communication paths.

OSS DISTRIBUTED ENTERPRISE BENEFITS

OSS DISTRIBUTED ENTERPRISE COMPONENTS

OSS DISTRIBUTED ENTERPRISE SUCCESS FACTORS