Clouds roll in to handle stratospheric capacity needs

By John Andrews
10:48 AM

With PACS and other diagnostic imaging files quickly diminishing healthcare data storage capacities, providers are scrambling to find a much larger repository to handle their needs. More and more, that means gravitating toward the cloud, IT vendors say.

Massive data generation, along with requirements for accessibility and long-term storage capacity, is necessitating healthcare’s migration to virtual storage in the many cloud systems available today. Russ Kennedy, vice president of product strategy and marketing for Chicago-based Cleversafe, says the astronomical amounts of healthcare data make cloud storage the only choice for providers. Kennedy cited a 2008 International Data Corporation digital study on the amount of data out there and determined that there were 800 exabytes at that time. One exabyte is equal to one million terabytes. In that study, IDC predicted that there would be 35,000 exabytes in the total digital universe by the year 2020.

Healthcare’s share of that amount, Kennedy said, is currently 500 petabytes (with one petabyte equaling 1,000 terabytes), and that will grow to 50 times that total by 2020.

“Providers now need a storage methodology that can accommodate this explosion of data,” Kennedy said. “The RAID (redundant array of independent disks) method is limited and can no longer handle it.”

Conversely, Cleversafe’s cloud storage system uses information dispersal algorithms, a process that slices data into unrecognizable pieces and disperses it across multiple locations.

“By using this kind of encryption and secret sharing, we’re able to create a cost-effective way to safely secure massive amounts of data at the highest level,” Kennedy said. “So not only does this system allow providers to prevent data leaks and solve security issues, but they can also and eliminate any unexpected outages which saves a ton of time, money and headaches in the end.”

Yet the expanded capacity of storage devices also presents a new type of risk, said Fadi Albatal, vice president of marketing for Melville, N.Y.-based FalconStore Software.

“The problem of data loss is massively amplified,” he said. “Now we have three terabyte disks available in storage systems, which means the loss of one hard drive could result in the loss of massive amount of records – an extreme cost to any healthcare provider. Healthcare institutions can no longer back up and recover the data in a timely manner. Even in a RAID protection set, the time it will take to rebuild a lost drive is very disruptive to any business operation. All these challenges are pushing storage providers, like FalconStor, to think of new ways to manage the data and the infrastructure that is hosting it.”

Many of the technologies available today, such as data deduplication, are deployed as stand-alone systems and will make their way into the primary storage controllers, Albatal said, predicting that further automation of data movement and migration based on data value and relevance also will start becoming a common function of the storage infrastructure.

“This in turn will aid in the adoption of newer technologies such as solid state to improve storage performance and will reduce the time and cost associated with IT infrastructure management in general,” he said. “However, the challenge remains in the separation between data storage and data management.”

Jim Champagne has seen the need for massive data storage coming since 1999 and guided his company InSite One into the diagnostic imaging storage space.

“We built a model that would be scalable over time so that it would have a high level of performance and security,” said Champagne, who is now executive director of Dell Healthcare Services after Dell acquired his Wallingford, Conn.-based company last year. “What we assumed in ‘99 is happening rapidly today. The cloud helps with storage simplification and virtualization. File sizes are growing and tools to use data like CT, pathology and genomics are emerging and retention periods are growing as well. HIPAA is a big part of it, along with security, disaster recovery and continuance. This model has been around for 10 years and is working very well.”

‘Glaring’ problems

“Healthcare is affected by the ‘soft’ costs of manual data protection, along with the issues of adapting new technologies into the environment,” said Pearring. “The enterprise nature of the healthcare business means it has a vast array of multiple platforms, which is very difficult for a single vendor to manage.”

Meanwhile, health system expansions and hospital acquisitions will continue to complicate <a href="/directory/interoperability" target="_blank" class="directory-item-link">interoperability issues as well, says Mike Leonard, senior product marketing manager for healthcare, Boston-based Iron Mountain.

“The growth of the volume of electronic medical records, medical imaging, and other medical information continues to accelerate, coupled with the need to share data for improving clinical services, operational efficiencies and accelerating revenue management,” he said. “The consolidation of data silos and linking the systems in which they reside, both across departments and through mergers and acquisitions, are all major contributors to the data storage challenges healthcare organizations face.”

As the healthcare industry works to get a handle on its data proliferation and storage capacity issues, they will face another major information wave with the arrival of ICD-10 codes in 2013, notes Clark Swainston, public sector and healthcare hardware specialist at Piscataway, N.J.-based SHI International.

“ICD-10 challenges our healthcare customers to bring forth complete coding of everything from medications to symptoms to procedures by 2013,” he said. “This only adds to the vast amount of data they need to support in their organization. This data needs to be stored and secured in a HIPAA-compliant solution.”

More regional news

CMS Administrator Seema Verma and National Coordinator for Health IT Dr. Donald Rucker

Is synthetic data the key to healthcare clinical and business intelligence?

The open source synthetic data source, Synthea. (Diagram courtesy of The MITRE Corporation.)