QoS Boosts Health of Telemedicine NetworkRenown Health monitors quality of service on its network to ensure high-quality telemedicine, video and telephony service for patients in rural areas.
Renown Health is the largest healthcare provider in Reno, Nevada, but it also serves patients in rural communities that may not have access to a large hospital or particular specialists. Renown Health realized it needed an effective means of reaching out to patients, so it created a regional telemedicine program that supports 45 communities in northern Nevada, Lake Tahoe and northeast California.
"We really wanted to operationalize a sustainable telemedicine program," says Renown VP Kirk Gillis. "Because our medical facilities were so far removed from rural patients, there was a tendency for patients in these areas not to seek early intervention for medical conditions, or to pursue preventive medicine."
- IT Budget: Getting A Seat At The Table
- Data-Driven Insights Help Healthcare Organizations Achieve Outcomes that Matter
White PapersMore >>
- Strategy: How Existing Security Data Can Help ID Potential Attacks
- Research: Beyond IT Service Management
Renown started its telemedicine initiative by investing more than a quarter of a million dollars in hardware, software and services for its network. This network had to be reliable, available and resilient, so Renown made a commitment to quality of service (QoS) to support its operations.
Those operations include transmitting a variety of medical information such as EKGs, ultrasound images, X-rays and more. It also had to support live video links between sites and VoIP. Renown has four hospitals, and also offers health services at 16 medical groups and eight urgent care centers in a 17-county region.
Renown went to Charter Metro Ethernet to get Ethernet virtual private line (EVPL) service to address network response levels and response thresholds and connect small urgent care centers with 10-Mbps circuits. "We also had a dark fiber ring around the Reno metro area locations and a couple of sites that we were running at 100 Mbps," says Clark. "We still used T1 in a couple of places where the metro Ethernet service wasn't available."
With the help of an outside consultant, Renown also implemented QoS policies for the network that included the handling of latency for the arrival of voice and data packets over the network and best case traffic prioritization to optimize network bandwidth and prevent bottlenecks and packet drops. "We wanted total, real-time visibility of everything that was going on in the network," says Renown Senior Network Analyst Val Clark.
During the rollout, Renown conducted extensive testing to make sure all of its QoS policies were working properly. "We would fill up the pipe with random packets and see if the service we were testing would be sustained with the QoS rules we had in place," says Clark. "These tests were very important because we had promised our users and executives that we would not go live with the network unless we could guarantee quality of service to all of the areas that telemedicine and other network services were going to serve."
Renown also was concerned about network endpoint management and being able to monitor real-time and historical network performance. End to end, the organization has 4,000 computers and 400 network switches.
Renown installed two GigaStor Observer units with network probes from Network Instruments. The probes could be "attached to network devices of our choosing and that we could readily scale with more probes as we expanded network coverage," says Clark.
Renown can get instantaneous reports on network performance from the probes and can look at historical information for trend analysis. "Before moving to a network probe concept, we really had no way of confirming for certain that everything on the network was working as it was supposed to, and the last thing we wanted was a false sense of comfort with how things were going," says Clark.
"With the analysis that is now part of our toolset, we can look real time at a stream of data for a videoconference, and we can immediately see if there are points in the network that are generating latency--and how we can mitigate it."
This is important since the packet capture from some of its applications is relatively small, which makes it hard to know how much they are impacting bandwidth overall, according to Clark.
"With QoS guidelines in place and with tools that now enable us to take a closer look at network performance at various network vantage points, we're not guessing anymore," says Clark. "Upper management certainly feels better."