A systems engineer shares his perspective on keeping track of college IT.
INTERVIEW | by Victor Rivero
Damian Solodow has been with Harrison College for over five years. A systems engineer, he came in at a time when the IT department was going through a large expansion to increase the role of IT at the college and expand the capabilities of the department. Previously, he has worked in server and application administration for over 10 years now and has been working with Windows Servers since the NT 4.0 days. Just before coming to Harrison College, he spent time at some software development companies as well as the local public library system. “My past experience in multi-vendor environment, especially at the library, helped to prepare me for the fast-paced, heterogeneous atmosphere here at Harrison College,” he says. Here, Damian talks about the network, applications, servers and monitoring it all to a good result.
Victor: How big is the network at Harrison College? How many critical applications do you monitor?
Damian: Harrison College serves over 5,000 students throughout the United States and internationally. Including the 14 campuses throughout Indiana, Ohio, North Carolina and online at Harrison.edu, the institution grants Associates and Bachelor’s degrees across five schools of study: Business, Health Sciences, Information Technology, Criminal Justice, and Veterinary Technology, as well as its culinary division, The Chef’s Academy.
The IT admin team manages approximately 170 servers, storage devices and appliances, both physical and virtual in a multi-vendor hardware environment. We use mostly VMware and XenServer hypervisors, with Windows servers as well as Linux and mostly HP Proliants and Dell EqualLogic SAN hardware.
Currently, we monitor 32 critical applications – mostly MSSQL, Exchange, Citrix XenApp, MySQL, Altiris, Backup Exec, SharePoint, CampusVue, Dynamics GP, with multiple instances of each of these or multiple servers running them.
Victor: What problems have you faced when it comes to application performance monitoring? Why is this technology important?
Damian: In the past, Harrison College used a rudimentary application performance monitoring technology solution, but found it to be limited in many ways. The product was simply not robust enough to serve as the comprehensive solution we needed for our dispersed network.
After implementation, we found that the solution lacked out-of-the-box metrics, thresholds and dashboards for each application, making monitoring and pinpointing issues quite difficult. We had to customize everything in terms of figuring out what metrics to monitor and what the thresholds should be for every application, including common out of the box ones.
We decided to try an alternative solution that could automatically monitor and pinpoint issues for our critical applications. We were already looking into part of SolarWinds’ IT Management suite for monitoring our network devices, and SolarWinds Server and Application Monitor (SAM) seemed a natural fit.
Victor: What have you learned from your past issues? What do you do now to track metrics, thresholds and dashboards for each application, monitor and pinpoint issues and handle physical and virtual servers?
Damian: We learned that we couldn’t compromise on our monitoring needs, given the size and critical nature of our network, but we also didn’t want a tool that required extensive customization or integration to achieve our goals. In the end, we found exactly what we needed – a simple-to-use solution that can monitor all our servers – both physical and virtual as well as all the applications across our campuses. We also discovered that there are some out-of-the-box features that we can’t live without.
Currently, we use SolarWinds’ Server and Application Monitor (SAM) to monitor our applications, track metrics and find issues. SAM makes the monitoring process simple and efficient by offering ready-to-use templates and provides the flexibility to create new, or customize existing templates. As Harrison College is a multi-vendor environment housing networking and storage devices from HP, Dell, F5 Networks and Cisco, we rely on SAM’s hardware monitoring functionality to achieve visibility into the status of all our hardware under a single monitoring umbrella.
In the two plus years we’ve been using SAM, we’ve created various logical service groups of application categories and remote campuses for quicker identification of issues. Also, we’ve created a specific template to monitor each third-party vendor’s application that sends application performance alerts directly to the concerned vendor, reducing operational time and communication between the college and the vendor in reporting the issue.
Additionally, we use SAM’s virtualization monitoring capabilities to gain greater insight into our virtual infrastructure, and monitor availability and basic performance metrics including CPU and memory utilization for each VM.
Because of the way our network is set up, SAM typically alerts us to warning signs or issues before our users feel the impact. With our past monitoring system, we were often finding out about application issues when calls started coming into the help desk alerting the staff to applications not functioning properly or outages, then we’d go about working to fix them. With SAM and the rest of our monitoring system in place, we typically receive information on issues before users realize there is a problem allowing us to have a solution in the works before they even call the help desk or notice that there is an issue. This proactive approach to monitoring has allowed us to really reduce the impact of outages on our staff.
Victor: Got any good stories that illustrate how your application monitoring system has prevented or solved problems on campus?
Damian: Since implementing SAM we have realized an increased level of stability on our network, so the frequency of issues has declined and there aren’t as many stories to tell.
Thinking back about two years, we lost power to our datacenter because of a car accident that took out a utility pole and cut the power to a three block area for about four hours. Once the utility workers had the power restored, SAM made it easy for us to see that all of our applications came back online quickly and correctly. When you have an outage like that there is usually a great deal of scrambling around to make sure you’ve got everything back online and SAM provided our team with a single pane of glass view to monitor all of the applications and their status. It really decreased the stress and headaches involved in a situation like this and made it much easier to make sure the power outage issue was behind us. The view, and the solution itself, helped us feel confident that the problem was resolved correctly.
Victor: Any thoughts about technology is transforming IT management in education?
Damian: I’m seeing a lot of technologies coming into play recently that help us in IT departments for schools take a more proactive approach overall. Instead of acting like a fire brigade, running around putting out fires, it allows us to deal with things proactively and simply so that we can focus on helping technology better serve the school, staff and students.
Additionally, educational IT seems to be a “late adopter” when compared to the enterprise. The space it seems to be shifting into now, from a technology and IT perspective, is where the enterprise was about two years ago. The major barrier to entry to some of these newer capabilities – cost – has become lowered due to new, easy to implement technologies on the market. It’s making it easier for educational IT to shorten that adoption period and keep up with technology trends.
Victor Rivero is Editor in Chief of EdTech Digest. Write to: [email protected]