ARCHIVED: Completed project: Confio Ignite Oracle and SQL Server database and VM performance monitor

This content has been archived, and is no longer maintained by Indiana University. Information here may no longer be accurate, and links may no longer be available or reliable.

Primary UITS contact: Dan Young

Completed: July 2, 2012

Description: Historically, the Enterprise Database Administration team has been engaged by various departments and divisions in database diagnostic and tuning assistance for enterprise applications. These requests are often in response to critical production issues and can require extraordinary multi-team efforts to reach resolution. Ignite is an Oracle and SQL Server performance monitoring tool that will enable both Enterprise Infrastructure and Enterprise Software teams to more effectively and efficiently respond to critical production issues, as well as proactively monitor and alert to limit the frequency and the impact of critical production issues.

Outcome: Once complete, the implementation of Ignite will allow for unified monitoring, analysis, and tuning for both Oracle and SQL Server databases and their applications. This will allow both infrastructure and development groups to proactively monitor, research, and troubleshoot performance issues, and should also help with future planning. Additionally, this should help bridge the functional gap between EI and ES as regards application performance issues.

Milestones and status:

4th quarter 2011: SQL analytics software research, evaluation, and review
January-May 2012: Negotiations, funding proposal, and formal software acquisition
June 2012: Implementation of production Ignite server and repository
July 2012: Release of Ignite to end users

Comment process: Email Dan Young.

Benefits:

Agentless low-impact SQL monitoring for both Oracle and SQL Server environments via an intuitive web-enabled user interface
Detailed performance data, informing developers what portions of the SQL code base are performing acceptably and what portions are candidates for remediation
Performance recommendations and advice, which may be incorporated as part of a feedback loop into the development process, enabling more efficient coding techniques
The ability for ES to self-monitor and notify on critical SQL statement behavior
A longitudinal view of RDBMS system health, both individually and on the whole
Data to form baseline performance metrics on a server-by-server basis
Measurements of resource utilization, enabling informed decisions regarding the hosting/consolidating/expanding of database systems based on performance data
A view of enterprise-level resource utilization that is integrated with database resource consumption
Availability of historical data for performance comparisons and performance research
Reduced diagnostic times in relation to underperforming SQL and degrading performance trends

Primary client: UITS Enterprise Software and Enterprise Infrastructure

Client impact: The primary impact will be increased efficiency and productivity when confronted with performance issues, specifically as it relates to identifying and remediating underperforming SQL.

Project Team: The Enterprise Database Administration team

Governance: John Weakley, Director, Enterprise Infrastructure