One of the most exciting parts about the IT community’s annual Interop Las Vegas conference is that it gives the PathSolutions team the opportunity to meet face to face with customers who are actively using PathSolutions’ TotalView suite to streamline network troubleshooting in their data centers. It’s always a treat to hear client stories about how the software improves daily network conditions and makes life easier for the IT professionals tasked with ensuring strong system performance.
This year, PathSolutions’ CTO Tim Titus was able to sit down with Brian Chee, a PathSolutions’ customer and scientist from the University of Hawaii’s School of Ocean and Earth Science and Technology, for a “Live From the Show Floor” technical discussion.
Chee’s PathSolutions use case is a bit different from what the majority of our customers experience. That’s because his data center operates about three miles below the surface of the Pacific Ocean, at the University of Hawaii’s Station Aloha Cabled Observatory. Unlike other customers who operate on land, Chee and the Station ALOHA team must ensure uptime with more than 7,000 pounds per square inch of pressure bearing down on them as well as maintain a great length of sensitive underwater cable.
Making matters even more complicated is the fact that research money is scarce and operations are expensive, with ship rentals costing about $30,000 per day. And deep-water remotely operated vehicle (ROV) rentals run upward of $40,000 per day on top of that. As Chee explained, it’s an abhorrently expensive endeavor; just cable pressure housing, for example, costs hundreds of thousands of dollars.
As a result of these high costs, as well as the risks of conducting research in a high-risk environment where just a few minutes of downtime could ruin an experiment, it’s difficult to offer guaranteed service-level agreements to researchers that want to use the equipment.
“People want [the equipment] to work,” explained Chee during the interview, talking about the need for advanced protection. “They want their applications to function; they want to be able to do their jobs. They don’t want the network to stand in the way.”
Chee went on to explain how he is able to gain the necessary visibility and data to easily perform real-time network troubleshooting using TotalView. In fact, it’s so easy, non-technical PhD students working in the facility—that is, people with little to no knowledge of how networks operate—are able to use PathSolutions’ software to identify the cause and location of system errors. This allows Chee to use his high-paid technicians to concentrate instead on difficult projects like installing custom drivers, and building storage-area and software-defined networks. This helps him use his scare resources more effectively.
What are some examples of how PathSolutions is helping Station ALOHA? In one instance, the software discovered a misconfigured device that, in the words of Chee, “was spewing all kinds of garbage onto the network.” The team was able to turn off the machine and fix the situation before it caused any complications.
“More recently,” added Chee, “We found a Linux-based machine that had two Ethernet interfaces that we were bridging together. Ethernet 0 was on 100 MB, Ethernet 2 was on 100 MB, but the bridge was at 10. So we had a goofy bottleneck. I found that right in the weather report from PathSolutions.” PathSolutions’ Network Weather Report, in case you’re not aware, is an automatically generated report, emailed every 24 hours, that provides a detailed analysis of interface health and stability.
The chief take-away from the interview? The design of the PathSolutions’ TotalView network monitoring suite, which uses lightweight coding to scan networks and report errors faster, makes life easier for IT professionals—and people are starting to take notice. This is why PathSolutions has graduated from a standard 10-by-10-foot booth at Interop to a much larger 30-by-20-foot booth.
“There are thousands of network engineers out there that are getting more sleep now,” said Titus during the interview. “They aren’t getting woken up at 2 a.m. anymore (to fix network errors) and they are able to have a better lifestyle.”
Want to learn more? You can access the full interview by clicking here.