High-performance computing (HPC) acquires a large amount of data for complex operations through extremely fast processing speed, real-time analysis of data, and the goal of rapid decision-making. HPC is widely used in oil and gas exploration, weather analysis, gene sequencing, university scientific research, and cutting-edge laboratories. The storage system requires sufficient performance to ensure intensive and diverse analysis behavior from multiple HPC servers. Since the unanalyzed raw data will accumulate more and more data will be studied / processed in the future, capacity and scalability are also important considerations. HPC's overall data will eventually reach the PB level, requiring huge storage capacity to complete the archive.
EonStor CS is a scale-out NAS storage system. Up to 144 nodes can be added to form a high-performance large-capacity cluster with read / write speeds up to 100+ GBps, fully capable of running high-performance analysis applications. More than 100PB of storage space for analysis data archiving. Performance and capacity are unified under a single namespace, so that multiple HPC servers can analyze and share data in the same folder / path, reducing the complexity of data access. CS also supports automatic load balancing and evenly distributes data from all nodes in the cluster. When a large number of users access data at the same time, the system distributes the workload through parallel processing to solve the performance bottleneck problem of a single node and effectively improve access efficiency.
HPC applications require long-term, uninterrupted analysis of multiple servers. If the storage system fails, there will be problems in the entire analysis process, resulting in service delays. Through multi-layer data protection of hard disks and nodes, the high availability of EonStor CS ensures that the analysis process will not be interrupted. If any link fails, the data will not be lost and the performance of the analysis will not be affected. If the hard disk is damaged, Infortrend's unique RAID technology ensures that the system maintains normal operation at the same performance level. In terms of node protection, copy and erasure code protection functions avoid downtime caused by a single failed node. If a node failure is detected, CS will initiate a self-healing function to recover data from the failed node. 
The topology of HPC applications usually includes a large number of storage nodes. The complex architecture requires a simple method to build / manage the entire storage cluster and reduce the management burden. EonOne provides a setup wizard to help user administrators easily set up storage and networking. Deploying the EonStor CS cluster through EonOne is very convenient. It takes only 5 minutes to complete the setup and only 30 minutes to complete the deployment of the cluster. In addition, EonView developed by Infortrend is applied to Windows and Mac platforms, and provides intuitive and streamlined operation wizards. Even non-professional IT staff can quickly achieve data sharing and backup. After starting EonView, you can automatically search for storage devices in the terminal's network segment, and you can quickly mount shared files, and realize regular data backup through folder schedule backup and link recovery operations.