Pure Storage Releases FlashBlade for Clinical Data Storage

January 25, 2017 - Pure Storage announced the general availability of its FlashBlade solid-state array storage appliance, shipping in 8.8TB and 52TB capacities, to assist with clinical data storage.

FlashBlade was originally announced in March 2016 and has been available in Directed Availability since July 2016. Early adopters were able to build uses cases in several verticals including healthcare.

Pure Storage aims to provide organizations with a storage platform that can quickly deliver massive amounts of data in the face of IT infrastructure digitization. Organizations can collect more data with mobile, Internet of Things (IoT), and connected devices. With all the data coming in, organizations need a storage solution that can physically hold the data and one that can quickly sort and retrieve the data.

FlashBlade is built for handling high performance, unstructured workloads that are common with organizations transitioning to digitized infrastructure.

Advanced infrastructure solutions generally require organizations to increase storage requirements. When entities adopt several new solutions quickly, storage may become an issue if it isn’t addressed before implementation.

IDC Research Director of Storage Eric Burgener said in a statement that as more organizations are considering big data analytics, traditional storage architecture is not well met for the amount of unstructured data collected.

"Big data flash platforms are optimized to handle very large unstructured data sets with high degrees of concurrency while delivering flash performance and reliability,” Burgener concluded.

FlashBlade is an elastic scale-out system, optimized for high concurrency, bandwidth, and input/output operations per second (IOPS). This allows it to support more demanding workloads and high performance applications. It delivers an all-flash performance to petabyte-scale data sets and is comparable to legacy hybrid arrays.

An all-flash architecture is less volatile than a hybrid array because it does not contain spinning hard disk drives (HDDs). The flash can transfer data between solid-state drives much faster making it elastic and easy to scale.

During Pure Storage’s Directed Availability preview last July, UC Berkeley EECS department deployed FlashBlade to help with their ADAM/Big Data Genomics project. The department builds healthcare tools and technologies, altering them to move from academic tools to clinical care tools.

UC Berkeley researchers deal with massive genomic data sets that need to be visualized. They required a storage solution that was fast enough to retrieve and run data-intensive visualizations in three dimensions.

After the research is complete, the correlations and findings from the research can be used by medical professionals to make more accurate clinical diagnoses.

"When our researchers want to visualize a pattern or interaction, it creates a significant tax on our storage systems. Visualizations inherently require so much data to run, that they are nearly impossible to run smoothly," said UC Berkeley's Center for Computational Biology Professor Anthony D. Joseph. "FlashBlade removes that roadblock for our researchers, which means our findings positively impact public health faster. If a doctor knows exactly which treatment to prescribe a patient based on that patient's genetic makeup, that patient is going to recover sooner."

Healthcare organizations have been struggling with how to store unstructured data since the general adoption of connected medical devices. Many organizations are looking to big data analytics to qualify for federal funding, but may not have the storage capacity to support the volume of data with their current storage architecture.

Organizations need more than a qualified analytics solution to sort and connect data. A capable storage solution is the foundation of an analytics solution. Modern storage solutions using more advanced methods for data retrieval such as flash and solid-state drives are going to allow organizations to quickly make use of their data, getting faster results for real-time analysis and diagnosis.