By Doc D’Errico
Data could well be described as the DNA of modern healthcare. Providers are collecting more data than ever before and using it more strategically to open up new fields of research, adopt new approaches to disease treatment and pursue new strategies for the overall improvement of public health.
Their biggest challenge? Once they capture this data, they need to store it and manage it in an economical way. The entities that do the best job of facing down storage challenges that are growing exponentially, year by year, will be best positioned to succeed in the future.
Data volumes are swelling as a result of several factors – everything from technology advances like 3D imaging (estimated 8.5 times the data storage requirement of 2D imaging) to genome sequencing for the masses.
Disease prevention models like precision medicine rely on data to move from generic treatments to more precise, population-focused diagnostics. Precision medicine factors in data from patients’ genes, environment, lifestyle factors and family history, into clinical decision-making for earlier, more accurate diagnoses, and more effective treatment and prevention.
As the Internet of Things (IoT) becomes pervasive, more and more data is “machine”-created, coming from billions of sensors and other automated data collection points. This is especially true in healthcare, which is trending toward patient wearables and mobile medicine. Taken as a whole, healthcare data growth predictions eclipse those of many other industries.
Storing and protecting this mountain of data is pushing information infrastructure and budgets well past their limits. Yet, in the fast-paced connected world, simply storing and protecting data is not enough. This data must also be analyzed, manipulated and made available to an increasing number of people and systems as quickly, efficiently and cost-effectively as possible.
This puts pressure on the IT departments in healthcare organizations. Leadership is challenging them to find tools that manage all of this data in new ways to allow better and easier collaboration, portability, better workflows, and faster, lower cost access to information at the point of engagement. However, IT is not a core competency of most healthcare providers. It requires a huge investment in people, processes, and facilities today, and even more so for the future to make use of the data that is coming.
To meet their storage needs of the future, healthcare IT departments will need to move beyond “accepted standards” and think more in terms of “what is possible?”
For example, healthcare organizations tend to adopt risk-averse strategies, refusing to allow storage platform densities to exceed 1 petabyte per failure domain. This has translated into high data costs per gigabyte for their companies in the form of acquisition and management costs for multiple “islands” of storage, rather than taking advantage of a single, more inherently efficient, multi-petabyte platform.
Many healthcare organizations also have defaulted to the use of All Flash Arrays (AFAs) – external storage arrays that use only flash media for persistent storage. When first introduced, AFAs provided a solution for a specific workload performance problem – namely small, random reads for databases. But these Solid State Drives are not the best media for large, sequential reads or especially writes. They solved an issue in the data center, but at the same time, increased the overall cost of storage.
New storage innovations are needed
Storing the data that will drive innovation in their industry’s future is shaping up as one of the biggest challenges healthcare providers will face. Here are a few storage innovations that are needed to meet this challenge:
- Scalability – As healthcare IoT generates more data from mobile devices and an assortment of wearables, storage solutions need to quickly and easily scale to meet this increased volume.
- Security – With security threats on the rise, organizations need efficient, reliable encryption of all data sets to meet regulatory requirements without huge capital expenditures.
- Reliability – It is critical that storage solutions have the highest reliability, lowest risk of failure and lowest risk of performance degradation – especially when they’re managing sensitive health data.
- Consistently high performance – Speed is critical. Every half-second spent waiting for a medical record to be retrieved from a rolling nurse’s kiosk makes a difference in the quality and timeliness of service that a patient receives. To help healthcare providers deliver faster and better service, they need solutions with the fastest possible response times so they can spend less time waiting for data to be retrieved.
- Flexibility – Healthcare data volumes will continue to grow, pushing organizations to create solutions that allow for flexible expansion in a cost-effective way.
- Integration – Since healthcare organizations generally have multiple different applications and solutions in place from various vendors, it’s important to have solutions that easily integrate with their existing infrastructure.
The pressure is on Healthcare IT providers to innovate new products and services to meet this challenge. Often, what has been understood to be the “accepted standard” influences the possible solutions to a problem – obfuscating innovation that might shorten the path to a solution and create new options that were not previously possible.
As an example, to address the needs previously outlined, Infinidat has focused not just on media evolution, but on software. As the result of over 140 patented innovations, the Infinidat Storage Platform leverages a “Neural Cache” that uses predictive analytics through AI and Machine Learning to serve I/O requests from a small amount of the fastest media in the industry, while at the same time maintaining data availability through innovations in distributed data layout across an active:active:active hardware architecture. The combinations of these innovations provides industry-leading ROI and performance metrics for real workloads while transforming how Infinidat Customers enable their own competitive advantage at multi-petabyte scale.
Like in any discipline, in IT it’s the innovators who turn the statement, “What is” into the question, “What is possible?” The fact is that the “What is” for data storage today is simply no longer sufficient to meet the rapidly changing requirements of Healthcare Information Management. Or, to put it more succinctly, lack of innovation in data storage is putting limits on Healthcare’s “What is possible.”
Storage Innovation in Genomics Research and Cancer Treatment
Doctors and scientists at one of the world’s leading medical research centers are engaged in state-of-the-art clinical and pre-clinical research that provides cancer patients with the most advanced diagnostic and treatment support. The team is running a combination of Intel- and Nvidia-powered servers with multiple InfiniBox systems from Infinidat to store and process petabytes of genomics data. Similar to other genomics research organizations, the center stores and processes enormous amounts of genomics data, supporting a mix of FASTQ, BAM and VCF files.
Installed in 2016, the high-performance, high-capacity Infinidat solution supports sequencing, analysis, and data archiving at a cost that is significantly below alternative solutions, with zero storage-related downtime in four years. Recently, when the storage administrator was looking for temporary capacity for VMware datastores, he was able to borrow space from the InfiniBox to run this workload, in addition to the ongoing genomics research, without researchers noticing any impact to performance.
Storage Innovation for Improving Patient Outcomes While Reducing Costs
One of the largest regional health insurance providers in the U.S. was experiencing a 30% annual increase in data consumption, brought about by electronic health records, large volumes of medical imaging files, and clinical reporting requirements. Five years ago, the IT team concluded that the data storage industry simply wasn’t keeping up with the realities that were transforming their business. Today, the IT team can rapidly respond to new provisioning requests and leverage data analytics to improve member services and patient outcomes while reducing costs.
High-performance storage infrastructure from Infinidat has enabled a transition from business intelligence (BI) to predictive analytics, powered by AI and deep learning. Analysts now have real-time access to data reports that once took hours to produce. By harvesting insights from phone calls and pharmacy records, the organization can proactively address issues and trends that affect the wellbeing of its members. Data-driven insights also enable incentives for healthcare providers to reduce infection and readmission rates.
Doc D’Errico is VP, Office of the CTO, at Infinidat, an enterprise data storage company. Prior to joining Infinidat, Doc spent 25 years at EMC Corporation where he led key technology innovation initiatives. He holds 14 storage-related patents, a Ph.D. in Computer Science, and undergraduate degrees in Electrical Engineering and Business.