The healthcare industry, like all enterprises that deal with massive amounts of data, can find it difficult to balance the trade-offs between their need to maintain highly performant storage systems while at the same time limiting the costs for maintaining those big storage boxes on-prem.
NetApp has a new solution for AFF and SSD-backed FAS storage system users that the healthcare industry can use to its advantage: the new Cloud Tiering service. In this article we’ll take a closer look at this industry’s use case and see in detail how NetApp solutions including Cloud Tiering can address the challenges of healthcare data management, analytics, and research.
What Drives Data Usage in the Healthcare Industry?
Data is the fuel that goes into any industry’s engines, but in the healthcare industry data is much more organic—it’s the air that healthcare breathes. There are three main ways that storage is intrinsic to how this data is being used: clinical data management, AI and analytics, and genomics.
Caring for Data Means Helping Patients
Every time that a patient enters a hospital—and consider that for most people, entering into the world happens in a hospital—they begin to accrue a medical history. That data will need to be referenced every time that the patient visits the doctor’s office or emergency room over the course of their lives and those visits can be sporadic. For all the time in-between, the healthcare provider won’t need to use that patient data. Maintaining this data on performant storage systems such as NetApp AFFs is not the optimal situation.
For one thing, such storage systems, however performant they may be, are finite. At some point all that patient data will wind up consuming all the available space in the system. To scale, a healthcare institution like a hospital or medical research facility must then roll in additional storage boxes to expand their capacity, which can be quite expensive considering just the purchase costs alone, and even more so when taking into account their operating and maintenance costs.
Another important aspect of clinical data management is compliance. HIPAA is a powerful compliance law that requires that patient data be maintained for specific periods of time and that it is kept under the strictest privacy controls possible. The resulting fines for violations can be extremely harmful to operating budgets for medical facilities and companies alike. Making sure that data is retained properly and at as low a cost as possible is a crucial part of managing this challenge.
Building Better Diagnostic Tools with AI
Patient data is not the only concern that medical industry businesses have. Storage plays a crucial role in many of the most important groundbreaking work being done in artificial intelligence (AI) and machine learning.
AI and machine learning are changing how doctors diagnose illnesses, making it easier to detect things that are wrong and make the right decisions faster than ever before. To do this, AI and machine learning programs need to be trained through a deep learning process, which involves ingesting massive amounts of data. As the deep learning reviews data through continuous testing, it is able to adjust based on past errors and improve its ability until it is a honed tool that medical practitioners can use in live healthcare environments. AI can also consume large amounts of data if they are being used to process medical images, research and development data, and patient records.
The challenge is in efficiently storing all of that data used by those data-hungry applications, especially when using on-prem systems. Managing the costs for data storage is a key challenge for healthcare companies using AI.
Massive Data Sets for the Human Code
Genomics is the branch of molecular biology that is concerned with mapping and sequencing genes, most notably the human gene code. This information can be used to help prevent disease and to develop powerful new treatments.
The main challenge that genomics data usage presents researchers is two-fold: the data sets involved are extraordinarily large and sequencing genomics data takes a large amount of processing power. While some of the most basic parts of the genome have been made accessible in extremely compact files, there is still a huge demand for storage when sequencing genes. Not all of that data will be used all of the time, and that can lead to inefficient use of on-prem storage, which is ideal for this use case due to its performance demands.
NetApp in the Healthcare Industry
Consider how NetApp has already changed the healthcare industry through its storage systems. Take a look at some of the customer success stories for NetApp solutions in the healthcare vertical:
- McKesson relies on a combination of on-prem data and cloud solutions to keep their system available and scalable.
- JFK Medical Center utilizes all-flash NetApp storage systems to bring excellent care to communities in New Jersey, allowing them to lower costs and increase performance.
- PetaGene uses a number of different NetApp solutions, including FabricPool technology to send its archival genomics data to the cloud to reduce storage usage and costs.
- And more healthcare case studies you can read about here.
But healthcare, like every industry, is also moving towards a cloud-first model that is more scalable and based on an easier-to-absorb PAYGO spending model. Google Cloud, AWS, and Azure are all extending their specific solutions to the healthcare industry. But there is still a strong incentive for healthcare industries to retain on-prem data centers, and not the least of these reasons is the performance trade off. There is simply not the performance level in the cloud that users can achieve using systems such as NetApp AFFs of SSD-backed FAS systems.
Cloud Tiering Service: A New Solution for Healthcare Data Challenges
Cloud Tiering makes it possible for medical industry storage systems to focus on the important life-saving work that they need to do without worrying about running out of storage system space prematurely.
Cloud Tiering leverages the same data tiering technology that PetaGene employs for archival data to move data intelligently and automatically to an object store in the cloud. This object storage service can be in Amazon S3 on AWS, Azure Blob, and in more clouds coming soon.
Object storage offers a number of benefits that make it an attractive choice for cold data, the first of which is the price. This is one of the least expensive ways to store data but still keep it affordably accessible should that patient walk back into the waiting room with a new ailment. Clout Tiering will automatically detect that cold data is being read by your application and bring the data back to the performance tier on AFF or SSD-backed FAS for immediate use. Once the data reaches its cold threshold again, it will be sent back into the cloud.
This limits the amount of CAPEX spending that needs to be done and allows storage to be classified as pay-as-you-go, OPEX costs. Storage doesn't have to be in the same category of spending as purchasing new hospital beds and defibrillators, it’s essentially a utility similar to keeping the lights on.
For AI and genomics, the advantage is that the massive data sets in use no longer have to take up precious on-prem storage capacity. The data center storage unit can devote most of its ability to processing and hosting applications, while the data in the cloud is always ready should it be time to run new tests or review older builds.
Saving Space for Data that Saves Lives
In the healthcare vertical, it’s not just the health of the business that’s important, but also the people who need to be treated and can benefit from the latest medical discoveries and technology. NetApp has been a solid partner for medical institutions to help make those tasks easier, and the Cloud Tiering service for AFFs and SSD-backed FAS systems is the latest.