Feature | June 29, 2011 | Alan Howard, Director of Information Technology, Princeton Radiology

Princeton Radiology uses a fluid data environment to manage and store its data

The Dell Compellent data storage array stores some of the 64 TB of data Princeton Radiology keeps at its two sites

The Dell Compellent data storage array stores some of the 64 TB of data Princeton Radiology keeps at its two sites.


Coupling 50 years of experience with the latest radiological imaging technology and treatments, Princeton Radiology is a global leader in the diagnosis, treatment and research of disease. With five offices located across the state of New Jersey, our practice maintains millions of radiological patient records, including 20 years worth of electronic medical records (EMRs).

At Princeton Radiology, we are improving patient care and making data easier to share between offices and physicians. Digital images have changed the way our offices operate. These images provide a clearer end result, enabling our nurses and physicians to make diagnoses, provide answers and ultimately construct treatment plans for our patients much more quickly.

From an IT perspective, digital imaging is easier to preserve than film because it doesn’t age as quickly and stays viable for longer periods of time. Thus, the decision to transition to digital was an easy one. As we expanded our digital capabilities and the number of EMRs rose, we quickly realized our data storage capacity and availability couldn’t keep up with our growing needs.

Need for Storage and Availability Leads to Growing Pains
The company’s legacy storage system was hampered by this rapid data growth and we clearly needed a new storage system that lowered our total cost of ownership. We projected that the capacity limits of the legacy drives would soon render our system obsolete, and implementing a more efficient system with faster access to our data was a must.

It became crystal clear that we needed to make a change in order to store all of the EMRs for long- and short-term duration of the patients’ care and be able to take on new projects in the future. A new GE Centricity picture archiving and communication system (PACS) was one of the projects that required greater system storage capacity to accommodate newly created digital images.

Sharing patient information with hospitals and medical practices requires the ability to make that information available instantaneously, and in turn requires immediate data recovery in the event of an accidental loss or deletion. Our legacy system struggled to recover and protect patient data through the inefficient 12- to 14-hour recovery process it took to restore from tape. Recovery times for images and medical records lagged, offering us spotty reliability when we needed it. The best-case scenario for our backup recovery point objective (RPO), the length of time required to restore the system after a disaster, was 24 hours. For a provider in the healthcare industry, this was unacceptable. Princeton Radiology’s staff requires access to patient data 24/7 in order to provide premium care to our patients.

Legacy System Unable to Adapt
At Princeton Radiology, we were facing data accessibility challenges that many other hospitals are now coming to realize. The process of data management grew as a concern for our legacy system because patient data has such a long lifecycle and can become inactive, or infrequently accessed, over extended periods of time. Patients can come and go, but the electronic data and digital images that are part of their electronic patient records need to be accessed quickly, regardless of the frequency of patient visits.

Patient data records need to be readily available, as physicians have limited knowledge on which records they will need on a day-to-day basis. With a wave of data flowing into the new PACS, and given these accessibility challenges, Princeton needed a solution that could store patient data cost-effectively, while keeping it online for quick and easy access.

In March 2006, we identified a solution that allows us to efficiently accommodate our burgeoning storage growth and provide cost-effective data protection in the Dell Compellent storage area network (SAN). Our primary applications currently include a radiology information system (RIS) and a single integrated PACS, which was formerly two separate systems based on Microsoft SQL Server databases. In all, we are hosting approximately 64 TB of information: 37.5 TB at our primary site and 26 TB at our disaster recovery site.

With the Dell Compellent SAN in place, we were able to address a number of our IT pain points. As we implemented the GE Centricity PACS, we realized that we could not predict precisely the amount of storage we might need over the long term. But with Dell thin provisioning, we could add and properly allocate more disk capacity quickly and allow our hospital to run smoothly, without downtime and without spending hours each day on data management. Thin provisioning allows our IT staff to scale capacity, connectivity and performance incrementally to accommodate 500 GB of data growth per month, without affecting the daily work of our medical staff.

Automated Prioritization of Patient Data and Faster Data Recovery
As our data grew, Princeton was still wrestling with the need to balance disk performance and cost. We needed to prioritize usage of high-performance storage in the most affordable way, so we implemented automated tiered storage to dynamically manage the movement of data across three different tiers of storage.

At any time we can quickly access our data, even if it has not been previously touched in months, while significantly reducing our reliance on the fast, yet expensive, disks. Implementing the Fluid Data architecture ensured that our RIS and PACS applications could run more quickly so our doctors could retrieve and review important digital images without waiting, while maximizing our storage efficiency and cost savings.

Since we implemented the SAN, 90 percent of our data has migrated from tier-one down to tier-three disks, without manual intervention. With the SAN, we can establish the policies and then let the software move our data automatically.

This automated data movement has saved us a tremendous amount on storage costs by allowing us to only add relatively inexpensive tier-three disks as the organization grew. We have not had to purchase any more tier-one disks in the five years since we implemented the system.

To protect these tiers of storage and improve data accessibility, we needed a backup and recovery method that was flexible and fast. With our legacy system, we could only back up our data once per day, and we had to be sure that no one was using the system at the time, leaving our hospital further exposed to data loss and creating disruptions for our staff.

Dell Compellent’s replay technology allowed us to dramatically improve data protection by taking frequent, space-efficient snapshots of data. These snapshots only capture changes to data stored on the system, which not only saves us disk space, but also recovers lost or deleted files in about 5 minutes, eliminating the need to resort to time-consuming tape backups. In addition, we replicate these snapshots to a remote site configured with low-cost storage, which provides us with affordable disaster recovery as a key part of our business continuity strategy. Altogether, our patient data is now more accessible for our physicians and nurses, and they can work uninterrupted while we continue to provide data protection for images and medical records.

Fluid Data Architecture Accommodates Rapid Data Growth
With the new system, we can affordably manage our data while spending only a few hours per week on storage administration. It allows us to do more with the current IT staff and avoid adding headcount as our data storage needs grow. We are able to manage our storage in-house and measure our growth more effectively than ever before. Additionally, we can now focus solely on our patient’s treatment without having to constantly monitor a growing storage issue for our medical staff.

By moving to a Fluid Data storage architecture, Princeton Radiology has been able to better accommodate rapid data growth and data availability concerns. Patient data can now be easily accessed even after a long period of time between visits.

Moving forward, we are better equipped to optimize costs to manage patient data and scale storage capacity for our digital images and medical records. Our data is more available and better protected, which ultimately allows us to serve our patient base more effectively, day in and day out.


Related Content

Videos | Information Technology

Industry trade shows and conferences seem to be making their comeback in 2024. And the Healthcare Information and ...

Time July 25, 2024
arrow
News | RSNA

July 23, 2024 — Professional registration is open for RSNA 2024, the world’s largest radiology forum. This year’s theme ...

Time July 23, 2024
arrow
News | Enterprise Imaging

June 28, 2024 — Konica Minolta Healthcare Americas announced today a strategic partnership with Apollo Enterprise ...

Time June 28, 2024
arrow
News | Artificial Intelligence

June 27, 2024 — RamSoft, a global leader in cloud-based RIS/PACS radiology solutions, and RADPAIR, a trailblazer in ...

Time June 27, 2024
arrow
News | Information Technology

June 21, 2024 — Konica Minolta Healthcare Americas announced a strategic partnership with Comp-Ray, Inc., a Christie ...

Time June 21, 2024
arrow
News | Radiology Imaging

May 31, 2024 — MDView, a virtual care and medical second opinion platform, surveys users of its Radiology Second Opinion ...

Time May 31, 2024
arrow
Feature | Digital Radiography (DR) | By Melinda Taschetta-Millane

Digital radiography (DR) continues to advance at a rapid pace with today’s technological innovations and evolving ...

Time May 06, 2024
arrow
Feature | Information Technology | By Melinda Taschetta-Millane

The Healthcare Information and Management Systems Society (HIMSS) Global Health Conference and Exhibition brought ...

Time May 01, 2024
arrow
Feature | Information Technology | By Jef Williams

The rapid growth of healthcare data has reached unprecedented heights, making up about 30% of the world’s stored data.¹ ...

Time April 30, 2024
arrow
News | Society of Breast Imaging (SBI)

April 11, 2024 — iCAD, Inc., a global leader in clinically proven AI-powered cancer detection solutions, announced today ...

Time April 11, 2024
arrow
Subscribe Now