QuantaStor High-performance Computing
For computationally intensive applications such as Monte Carlo simulations, CAE, Seismic Data Processing, computational fluid dynamics, optical design and 3D simulations, the QuantaStor SDS platform provides High-performance Computing (HPC) storage capabilities that scale with its unique storage grid architecture. QuantaStor SDS provides IT architects with the latest in storage software and hardware capabilities so that any HPC solution can be designed specifically to meet extreme workload and storage traffic needs. Leveraging open systems technologies and protocols including NFS, CIFS, SMB, iSCSI, and S3/SWIFT, QuantaStor SDS enables fast access to file, block and object storage.
Key QuantaStor SDS HPC Features
With support for the latest generation of 10/25/40/50 and 100GbE network interface cards and switches, QuantaStor SDS appliances support the fast interconnects required for HPC workloads by providing high throughput for storage traffic across HPC clusters.
Scalable All-Flash and Hybrid Storage
Divide up and move storage between network shares and pools of storage that are all-flash or a hybrid mix of SSD and HDD for high performance and high capacity storage. All-flash configurations are designed to deliver maximum IOPS and throughput numbers for mission critical HPC workloads and applications.
Using a combination of HDD with SSD for read and write performance acceleration, QuantaStor hybrid SAN/NAS configurations are ideal for HPC workloads.
With overall grid scalability to 64 appliances, QuantaStor SDS for HPC can combine any number of network shares into any number of unique Global Namespaces so that storage grids can deliver consistent performance, accessibility, and reliability as HPC environments scale. Different network shares can also be placed on storage tiers to make it easy for end users to find their storage resources across HPC clusters.
S3/SWIFT Object Storage
QuantaStor SDS object storage configurations enable organizations to deploy cost-effective, hyper-scale storage for HPC workloads using the S3 and/or SWIFT REST based protocols. Object storage provides data intensive HPC applications with the ability to scale to 64 petabytes and appliances across HPC clusters.