Close this search box.

While Working in the Lab…Late One Night: Avoiding the “Monster Mash” When Evaluating File Storage for Academic Research HPC Workloads

Authored by:

Within the scientific research labs at the top academic institutions, some of the world’s most famous discoveries have been made. Our universities are the hub of knowledge and discovery, in fact, 56% of our nation’s basic research is being conducted at universities.

Research projects across the US are rapidly producing exascale levels of small and large files, which need to be efficiently managed and safely stored. Researchers use high-speed simulation and compute methods to rapidly process data. The need to speed innovation and discovery is critical, so these organizations can’t be slowed down doing the “Monster Mash” with legacy storage in their data centers. They’re looking for modern file storage systems that can keep up with new technologies and data types, deliver reliable performance, massive scalability, and provide a path to the cloud to provision compute power, when needed.

Qumulo’s cloud-native distributed file system has been purpose-built to meet the performance and capacity demands of these academic research environments, in the data center and in the cloud. Qumulo’s modern file system delivers extreme low latency, IOPS, and throughput performance for today’s academic research HPC workloads, supporting mixed IO performance and handling billions of small files as efficiently as large ones. Qumulo’s single namespace eliminates silos with a single SMB share or NFS mount for NAS, home directories, instruments, processing and AI/ML. Data is consolidated and collaboration is accelerated.

Universities can benefit from Qumulo’s simple modular scalability to keep up with their continuous massive data growth and to support growing research projects. Qumulo’s software-defined architecture utilizes clusters of nodes made up of Qumulo hardware, or pre-qualified industry-standard hardware from HPE or Fujitsu. Managing growth is simple. Simply add nodes to the existing infrastructure to increase performance and capacity levels uniformly, when required, with no disruption or downtime. Qumulo’s scale-across architecture effortlessly expands performance and capacity across on-prem and cloud environments, including AWS or GCP, to burst compute performance, collaborate, or archive. Qumulo’s file storage delivers fast, consistent access to data across the entire system.

Qumulo’s real-time analytics eliminates data blindness, providing instant visibility across billions of files. This visibility is extremely useful for managing research workflows, as it allows organizations to gain control with real-time information about the entire storage system to predict usage, I/O hotspots, and capacity trends, and more proactively manage current and future requirements. System administrators can identify trends for chargebacks and apply quotas in real-time.

This Halloween, get out of your “Monster Mashup” of storage solutions, and take a look at Qumulo’s Thriller of a file system, along with our many platform options to meet your performance, management, archive and cloud requirements.

Related Posts

Scroll to Top