The analysis of the current situation at the Univerzal Bank Belgrade established certain deficiencies of the information system that were to be eliminated in order the system was brought to a level desired by the Bank and which follows the latest trends and standards set for a financial institution.

Central computers at which the databases were installed were four years old, operating in non-warranty period with low performance level. The two computers were installed as a single Cluster server based on Microsoft Cluster Services technologies. They had Windows 2000 Advanced Server operative system, which was also to be advanced to a newer, more reliable and better performing Windows 2003 Server Enterprise Edition.

Large  number of services of vital importance for the Bank’s operations and communication with other institutions (SWIFT, Central Securities Registry, Payments Operations, Western Union, Credit Office), were operating on already outdated hardware and software platform.

The disk subsystem was of capacity insufficient to support the growing needs of the Bank.

The central part of the communication network was insufficiently developed, without redundancy, with low-level performance and not enough scalable for introduction of new services.

There was not a disaster recovery scenario or equipment to enable implementation of such a scenario

Thus, the Univerzal Bank recognized the need to advance the existing services and introduce new ones. Therefore, the Bank opted for Pexim in order to acquire quality assistance and support in its restructuring project.

The basic goal of the Project was to establish production environment to enable availability of the Bank’s services on 24x7x365 basis using the advantages of the Microsoft Windows server environment, then Microsoft SQL base and Microsoft Cluster Services relying on EMC disk systems, IBM servers and Cisco network devices.

The project was finalized at the beginning of December 2006.

The following was delivered in a four-month period:

• There was a complete innovation of the Data-center of the Univerzal Bank

• State of the art serves were installed based on Intel platform. It is a matter of two IBM x3950 servers with 4 dual-core processors each and 16GB RAM. These servers are top-of-the-line of the entire range of Intel processors servers and are with high scalability level (they could be extended by up to 32 processors and  512GB RAM).The Microsoft Cluster was established on these servers,  at which Windows 2003 Server Enterprise Edition and Microsoft SQL Server in the Active/Passive alternative were installed. This enabled high-level availability, and in case of a node failure all severs automatically move to another node in less than a minute.
• All services of vital importance for the Bank’s operations were migrated to one IBM Blade Center. This platform enables the highest consolidation of server resources, where in seven rack units it is possible to place up to 14 IBM HS20 servers with two Quad-Core processors each. All servers were configured in order that the start up is done with subsystem disks, without having to keep the local disks.
• A new disk system was installed - EMC Clarion CX500, which has a sufficient projected space to satisfy the Bank’s needs in the next three years.
• The central network section was completely redesigned by introduction of a new redundant switch Cisco 4507R which greatly advanced the possibilities of the communication network and enabled introduction of new servers.
• The complete environment for Disaster Recovery scenario was established. At this stage, the Disaster Recovery was implemented in two different rooms in the same building while the plans are to relocate the entire DR equipment to another location, as soon s the location is prepared and functional. In respect of the DR scenario, the Mirror View EMC technology was used, which supports synchronous and asynchronous replication between two disk systems independent from the servers. The Bank opted for synchronous replication, which enables replication up to the level of the last transaction before an eventual failure of the system. The entire concept was successfully tested and proven in the practice. It should be underlined that the Bank opted for all vital services to be replicated at the DR location. Under this application, the Bank will be ready and be able to abide by the new legislative requirements for a DR location. It is necessary to only move all equipment to the future location and rent a link of relevant capacity.