Dartmouth-Hitchcock Medical Center (DHMC) provides healthcare services to thousands of residents of New Hampshire and Vermont. As the medical center became more dependent on technology to deliver high-quality healthcare, management realized that
its data center required uninterrupted availability. DHMC decided to build a disaster recovery (DR) facility that uses the Windows Server 2012 R2 operating system and Microsoft System Center 2012 R2 data center solutions to provide a low-cost, highly resilient
DR solution. The medical center will be able to drastically reduce downtime of critical applications that affect patient care, without needing the US$89,000 annual expense of a third-party DR solution. The existing DHMC IT staff can manage the new DR location
and a growing IT environment by using many new automated efficiencies, while enhancing their career skills.
Founded in 1893, the nationally ranked Dartmouth-Hitchcock Medical Center (DHMC) integrates high-quality patient care, advanced medical education, and research to provide a full spectrum of healthcare to the citizens of New Hampshire and Vermont. DHMC
includes New Hampshire’s only Level 1 trauma center and its only air ambulance service. In addition to its main campus in Lebanon, New Hampshire, DHMC has four major sites in Concord, Keene, Manchester, and Nashua, New Hampshire, and many smaller practices
throughout New Hampshire and Vermont. DHMC employs 10,000 hospital, clinic, and medical staff members.
DHMC has steadily increased its use of technology over the years to deliver better, more accurate, and more personalized healthcare services. The addition of more clinical and teaching applications, mobile devices, and communications and collaboration services
required adding more servers in the data center. DHMC was an early user of server virtualization software to curb physical server proliferation; specifically, it used the Hyper-V virtualization technology in the Windows Server 2008 R2 and Windows Server 2012
operating systems to reduce the number of physical servers by 70 percent and virtualize servers for most of its major clinical and productivity applications, including its 8,000-user Microsoft Exchange Server email messaging environment.
||By using Windows Server 2012 R2, we can get our entire virtual environment back online in one hour in the event of a disaster, drastically reducing downtime…. This is imperative because lives could be at stake.
| Robert McShinsky
Senior Systems Engineer, Dartmouth-Hitchcock Medical Center
By early 2013, the medical center had nearly 600 virtual machines running on just 36 physical host servers—all clustered HP ProLiant blade servers running Windows Server 2012 and Windows Server 2008 R2. It used Microsoft System Center 2012 data center solutions
to manage its data center infrastructure.
Keep IT Infrastructure Available
The more dependent that DHMC became on technology, the more critical it was that the underlying data center infrastructure be available around the clock. “Downtime of many of our computer systems seriously affects patient care and profitability,” says
Robert McShinsky, Senior Systems Engineer at Dartmouth-Hitchcock Medical Center. “With some applications, such as those used during surgery, lives could be at stake. Or if we couldn’t access key systems, we would have to turn away radiology patients needing
treatment. Although we’re in a rural location, we’re not the only choice that people have, so we want to deliver exceptional customer service all the time.”
To bolster the availability of its critical IT systems, the DHMC IT organization began laying the groundwork for a second data center that could serve as a disaster recovery (DR) facility. DHMC needed a technology solution that could provide complete replication
of the primary data center to this secondary location in a timely and cost-effective manner. McShinsky and his colleagues looked at storage area network (SAN) replication solutions, but they were extremely expensive and had some technical limitations. The
staff also looked at the Hyper-V Replica feature in Windows Server 2012 but needed a higher-volume solution. Hyper-V Replica, which replicates virtual machines over IP-based networks, could not rapidly fail over the hundreds of virtual machines that DHMC would
need to protect should the main data center fail.
Keep Costs Low
Although it was setting up a second data center, DHMC would not be hiring a second data center staff. It needed technology that would help its existing staff manage the second location by working more efficiently. “Even though we had virtualized extensively
and used System Center to create a private-cloud infrastructure that was highly automated, there were still routine tasks that took a lot of staff time and involved server downtime,” McShinsky says. For example, the process of upgrading to a new version of
Windows Server involved many manual processes, as did expanding virtual hard disk space when needed. When the IT team had to take a server offline for a maintenance task, it consumed time not only for the IT staff but also for application teams and even hospital
“We’re adding between 150 and 200 virtual machines to our data center each year, and anything that can help us administer this growing infrastructure without disrupting doctors and patients is a huge plus,” McShinsky says.
Create Attractive Work Environment
A modern IT infrastructure not only helps DHMC deliver better, uninterrupted patient care, but it also helps the hospital attract and keep IT staff. “Because we’re in a rural location, it’s sometimes difficult to find qualified IT staff,” McShinsky says.
“We need to rely on software to automate as many tasks as possible so we don’t have to hire as many people—and we also need to ensure that the staff members we do hire want to stay here.”
When Microsoft announced the Windows Server 2012 R2 operating system and Microsoft System Center 2012 R2, Dartmouth-Hitchcock Medical Center eagerly joined the Rapid Deployment Program (RDP) and deployed early-release versions of both products. When
the products become generally available, DHMC will use its Microsoft Enterprise Agreement license to upgrade at no cost.
“It’s really been uncanny; as soon as we need a new capability, it appears in the next release of Windows Server or System Center,” says Jason Weier, Systems Administrator at Dartmouth-Hitchcock Medical Center. “We’ve seen that happen again and again.”
Cloud-Integrated Disaster Recovery
This time, the DHMC IT staff saw its disaster recovery needs met by the newest version of Hyper-V Replica and the Windows Azure Hyper-V Recovery Manager technology. Hyper-V Replica has been enhanced in Windows Server 2012 R2 to support a third replication
site, to replicate hundreds of virtual machines at a time, and to support replication frequencies down to 30 seconds, which helps DHMC achieve nearly real-time replication at a much lower cost than fully synchronous replication solutions. Live Migration compression,
another Windows Server 2012 R2 feature, helps speed the process by greatly accelerating the transfer of virtual machines between host servers.
||Hyper-V Recovery Manager coupled with Virtual Machine Manager gives us an enterprise-caliber DR solution that is incredibly cost- effective and easy to use.
| Robert McShinsky
Senior Systems Engineer, Dartmouth-Hitchcock Medical Center
Hyper-V Replica works in concert with Windows Azure Hyper-V Recovery Manager, System Center 2012 Virtual Machine Manager, and Windows Azure to provide simplified, cloud-based management of the replication process. The control plane of Hyper-V Recovery Manager
is delivered as a cloud service that runs in Windows Azure (called the Windows Azure Management Portal) so that, in the event of a disaster, the DR orchestration metadata is securely accessible from any location.
“By creating a Hyper-V Recovery Manager Recovery Plan, we can orchestrate the DR process and replace most manual steps using scripts,” McShinsky says. “This dramatically reduces our recovery time in the event of a disaster.”
DHMC can also use Hyper-V Recovery Manager to test its DR setup, something that has been a problem with DR solutions in the past. Hyper-V Recovery Manager uses Hyper-V Replica and the networking support in Virtual Machine Manager to deliver simplified DR
drills with no impact to production workloads or the protection of those workloads. “DR drills are valuable for compliance reasons, for training our staff in DR procedures, and for verifying that our DR systems are working,” Weier says.
When the new DHMC DR facility comes online in the second half of 2014, the DHMC IT staff will have a centralized view of both data centers through the Windows Azure Management Portal. “Hyper-V Recovery Manager coupled with Virtual Machine Manager gives us
an enterprise-caliber DR solution that is incredibly cost-effective and easy to use,” McShinsky says.
DHMC plans to use Windows Azure as a third backup and DR location. To do that, it will use Windows Azure Backup to store data securely in a Microsoft data center for an added recovery option that’s much less expensive than renting or buying a physical infrastructure.
“It took us a long time to justify a second data center for disaster recovery, and it would be nice to have a third georeplicated location, but that’s probably not going to happen for budget reasons,” McShinsky says. “Windows Azure will be our third DR location.”
Windows Server 2012 R2 also features easier migration between Windows Server releases, allowing DHMC to seamlessly move workloads from Windows Server 2012 to Windows Server 2012 R2 with no downtime.
Phased Rollout, Big Workloads
DHMC worked with Microsoft Services during the Windows Server 2012 R2 RDP to configure and deploy the new software in the DHMC environment. A Microsoft Services consultant worked onsite with the IT staff to provide best practices for deploying Hyper-V
Replica and using the advanced networking capabilities in System Center 2012 R2 Virtual Machine Manager. “Our Microsoft Services consultant’s insights helped us learn these new features, especially Virtual Machine Manager,” McShinsky says.
DHMC has deployed Windows Server 2012 R2 Datacenter on several test clusters and plans to move into production with the new products by the end of 2013. By early 2014, DHMC will have the majority of its Windows-based workloads running in its Windows Server
2012 R2 private cloud. By the second half of 2014, its DR facility will be online.
“We feel completely comfortable virtualizing our biggest workloads, such as Microsoft SQL Server 2012, using Hyper-V,” McShinsky says. “We have hundreds of SQL Server instances that we’re going to virtualize.” DHMC plans to use SQL Server 2012 AlwaysOn and
AlwaysOn Availability Groups to provide an extra layer of disaster recovery for its database environment. AlwaysOn is an enhancement of database mirroring that supports as many as four secondary databases. An AlwaysOn Availability Group is a logical collection
of databases that automatically fails over as one unit. “Between the performance and scalability enhancements in Windows Server 2012 R2 and the availability features in SQL Server 2012, we can virtualize our biggest databases with no availability concerns,”
Going forward, DHMC expects to enhance its data centers even more frequently with an improved release cycle of Microsoft cloud software. “In the past, there was a lag between the release of new versions of Windows Server and System Center, which hampered
our ability to put the features of both products to work in our environment,” McShinsky says. “Now, Microsoft has committed to link the development and release cycles of these two products and to speed up the release cycle to every year, which means we’ll
have new improvements sooner.”
By upgrading to Windows Server 2012 R2 and System Center 2012 R2, Dartmouth-Hitchcock Medical Center gained a cost-effective disaster recovery solution that will help it deliver healthcare services without interruption. The IT staff is also using the
new software to reduce time-consuming maintenance work and associated downtime and to manage a growing infrastructure without adding staff. Using the latest data center tools makes staff members not only more efficient but also happier because they are continuously
advancing their skills.
Disaster Recovery Helps Ensure Uninterrupted Healthcare Delivery
Having a reliable DR solution is critical to the medical center’s mission as a healthcare provider as well as to its profitability. “Without many of our IT systems, we simply can’t see patients,” McShinsky says. “By using Windows Server 2012 R2, we can
get our entire virtual environment back online in one hour in the event of a disaster, drastically reducing downtime for applications that affect patient care. This is imperative because lives could be at stake.”
||Between the performance and scalability enhancements in Windows Server 2012 R2 and the availability features in SQL Server 2012, we can virtualize our biggest databases with no availability concerns.
| Robert McShinsky
Senior Systems Engineer, Dartmouth-Hitchcock Medical Center
Because Hyper-V Replica is built into the operating system and Windows Azure Hyper-V Recovery Manager and Windows Azure Backup are available through a Windows Azure subscription that DHMC already purchased, DHMC did not need to spend US$89,000 a year for six
years on a SAN-based replication solution for its Windows workloads.
“With Hyper-V Replica and Hyper-V Recovery Manager, we’re getting nearly the same functionality as a SAN replication system—30-second recovery windows versus full synchronous replication,” Weier says. “We need a DR site that can take over for our main site
in a short period of time, and Hyper-V Replica completely meets our needs.” Also, by using Microsoft software, DHMC avoids the complexity and support difficulties of a multivendor solution.
Automated Tools Reduce IT Costs
Beyond DR, the DHMC IT staff saves time every day by using Windows Server 2012 R2 and System Center 2012 R2 to automate and speed routine tasks. For example, the process of upgrading all of its Windows Server 2012–based virtual machines to Windows Server
2012 R2 will require zero workload downtime thanks to the easier migration between Windows Server editions. Moving 600 virtual machines to a new operating system edition previously would have required more than 600 hours of downtime, with most of that work
taking place after hours for IT staff.
Every time the IT staff wanted to expand virtual hard disk space for a virtual machine, it used to incur 30 minutes of application downtime, depending on the size of the virtual machine. With Windows Server 2012 R2, disk expansion drops to a two-minute task
that doesn’t affect users at all. Over the course of a year, this adds up to about 50 hours of downtime and IT administrative work avoided—not to mention coordinating the downtime and performing many of the expansions in early-morning hours to avoid patient
“We don’t have to involve the application teams in these infrastructure tasks, which would affect application users and sometimes patients,” Weier says. “You think that one 15-minute server outage is not a big deal, but it has a ripple effect on the organization
and takes a huge amount of time from many people and affects the quality of our services.”
Live Migration compression is another timesaver. Some DHMC host clusters contain more than 180 virtual machines. “Without compression, it could take several hours to complete cluster maintenance tasks,” Weier says. “Live Migration compression, paired with
the ability to live-migrate multiple virtual machines at once, allows for this maintenance time to be shrunk to less than an hour.“
McShinsky adds a strategic perspective on these daily efficiencies. “Our environment is growing, and we need every new feature that Microsoft comes up with, as soon as we can get it,” he says “Our management understands that data center efficiencies directly
affect the business by helping us provide great patient care and be more competitive. We’ve been able to use our Microsoft license to keep the innovation coming. As soon as the newest operating system version comes out, we have it. Microsoft software helps
us handle a growing environment without expanding our staff. We’ve gone from 400 to 600 virtual machines in just a couple of years and are implementing an entire DR facility without adding new staff. We could not have done that without Microsoft software.”
Modern Data Center Helps Attract and Retain Staff
A modern data center also helps DHMC IT staff members accomplish their jobs during standard business hours rather than extending into evenings and weekends. “With all the efficiencies we’ve gained, we can give IT staff a better quality of life, and by
keeping our data center up-to-date, we help them learn new skills,” McShinsky says. “It’s harder to find IT talent in rural New Hampshire than in Boston or even nearby Manchester, so it’s a draw to tell prospective hires that they’ll be working in a state-of-the-art
data center. Microsoft software is key to our ability to be innovative and attract top talent.”
Transform the data center
The hybrid cloud from Microsoft transforms the data center by extending existing investments in skills and technology with public cloud services and a common set of management tools. With an on-premises infrastructure connected to the Windows Azure platform,
you can deliver services faster and scale up or down quickly to meet changing needs.
For more information about transforming the data center, go to:
For More Information
For more information about Microsoft products and services, call the Microsoft Sales Information Center at (800) 426-9400. In Canada, call the Microsoft Canada Information Centre at (877) 568-2495. Customers in the United States and Canada who are deaf or hard-of-hearing
can reach Microsoft text telephone (TTY/TDD) services at (800) 892-5234. Outside the 50 United States and Canada, please contact your local Microsoft subsidiary. To access information using the World Wide Web, go to:
For more information about Dartmouth-Hitchcock Medical Center, visit the website at: