2. (6) The technology delivery systems are highly reliable and operable with measurable standards being utilized such as system downtime tracking or task benchmarking.
Not Observed Insufficient Moderate Use Meets Criterion Completely
All systems, both in-house and outsourced, are available 24/7 to students and faculty.
Systems are monitored during peak usage times to ensure sufficient stability. For in-house systems, monitoring software is used to track CPU and memory usage. If usage exceeds an upward threshold, event notifications are emailed to the SLIS system administrator. Excessive load situations are extremely rare, and typically involve false alarms generated by overly aggressive website crawlers. In such situations, steps are taken to block the misbehaving crawler via the Robots Exclusion Protocol, and efforts are made to contact the administrator of the offending system.
Managing Unexpected Downtimes Downtime of in-house systems is rare. In the event that downtime occurs, once the issue is resolved and availability restored, extensive efforts are made to document the issue(s) in order to prevent reoccurrence. Managers are notified in full detail. In most cases, end users are provided with a summary of the event.
For outsourced systems (such as Blackboard Collaborate and Canvas), in the event of unexpected service outages, the service providers will notify customers and provide event details. Where appropriate, the School of Information IT staff will disseminate some or all of the provided information to end users. For major outages, which are rare, the service provider will typically include a follow-up email to customers, often sent by a top-level manager, which includes a breakdown of the event and steps that have been taken to prevent a recurrence.
Tracking program growth to project and scale tech needs
The School of Information is constantly monitoring fluctuations in student enrollments, including projecting possible growths or reductions for upcoming semesters. All technology delivery systems are deployed with these fluctuations in mind, and most contracts with service providers have been negotiated with a margin for growth built into the licensing agreements. In the event of unexpectedly high growth, service providers typically include options for incrementally increasing licensing to accommodate the additional growth at a reasonable charge through an addendum to the original contract.
Scheduled maintenance on School systems is conducted during low usage periods, and where possible, is completed during the intersessions between semesters to limit impact on class activities.
If unexpected, urgent maintenance needs arise, all efforts are made to schedule the maintenance during non-peak hours.
For major maintenance windows, prior notification is sent to all students via an administrative mailing list to which all students are required to subscribe. All reasonable efforts are made to notify users of major maintenance windows at least 1 - 2 weeks in advance, and reminders are sent closer to the event.
For hosted services, such as Canvas, Blackboard Collaborate, and Blackboard IM, the service providers notify customers via email regarding scheduled maintenance windows. We in turn notify all students and faculty regarding the scheduled maintenance windows.
All in-house systems and databases are backed up daily. Once per week, all backups are written to tape. Tapes are then stored at a secure offsite facility.
For outsourced systems, service providers use industry standard data center practices for redundancy, backups, and failover.