Praxair, Inc. is an American worldwide industrial gases company. It is the largest industrial gases company in North and South America, and the third-largest worldwide by revenue.
Scope of Work
Praxair generates critical financial reports such as Trial Balance, Balance Sheets etc. where data integrity is assessed in real-time. KPI provides Cognos and OBIEE Reporting Applications support as part of the Managed Services package along with the ETL technologies. This demands highly skilled support with expertise in multiple tools such as OBIEE, Microsoft SQL Server Integration Services (SSIS), ODI, and Cognos BI Data Services. This requires a team which can work independently and communicate with the global user base – the team has to interact with users worldwide and support a development lifecycle that is reliant on collaboration between business and technical teams. The environment is in the process of constant redesign and migration of jobs from Cognos to ODI. The BI support team has to address these changes while stabilizing existing processes and improving performance. KPI supports ODI integration with financial data to many systems including Hyperion.
- ETL Production Load Monitoring
- ETL Performance Monitoring
- Production Load Failure Resolution
- Documentation and Email
- Data Model and Data Integration Enhancements
Oracle Business Intelligence 11g:
- Application administration
- Production support activities to address report data issues
- Enhancements and performance tuning of existing reports
- User and security management
- Automation scripts for some of administration tasks
Why Was KPI Partners Chosen?
KPI had worked with Praxair previously and were familiar with Praxair’s systems and architecture. KPI is a world leader in BI managed services and has rich experience in relevant technologies including Microsoft SSIS with JD
Edwards as the source system. Praxair selected KPI for their SLA-based approach, taking more direct responsibility for the client’s systems through proactive network monitoring and management.
- KPI provides in-depth knowledge of JD Edwards at Praxair
- KPI discovered duplicate records while diagnosing the root cause for a few repetitive failures and realized that they were caused by the violation of integrity constraints. The necessary clean-ups were done and corrective measures were taken to resolve the issues.
- KPI reduced the time taken to gather stats during ETL load by implementing a dynamic process to change the sampling percentage
- KPI identified data gaps caused due to bad ETL runs in the past and loaded the necessary historical data
- Performance tuning of Cognos cubes was done by proposing changes in the framework model, which reduced the cube runtime from 45 minutes to less than 5 minutes
- KPI has provided workload efficiencies, rapid ticket response, and overall service center reporting to allow the management at Praxair to view detailed performance reports
- KPI collaborates with Praxair on POCs, technology insights, and enhanced knowledge transfer
- KPI provided extended support for the financial close process and has become the main provider for all support allowing the Praxair team to focus on tactical projects
- KPI ensured service continuity including 24x7 global support for APAC, North America, EMEA, and IMEA using SLA-based control and expressive, on-demand service dashboards
- Proactive analysis and management of service tickets helped Praxair achieve a 28% reduction in monthly tickets pertaining to frequent user issues and improved the query performance of certain critical business reports by 80%. As a result, business users’ productivity and useful bandwidth increased by 7% on a daily basis.
- KPI automated the application data archival process and set up email alerts when the server ran out of space thereby improving the availability of key systems even during off-peak hours. As a result, IT systems became more responsive and business users had greater flexibility in managing their time.
- Reduced ETL failures by 20% and improved the stability of the overall process
- Reduced the nightly data load runtime by 40 minutes to 1 hour
- Reduced the average time taken to resolve ETL failures and production incidents
- Identified and addressed major integrity failures and data fallouts