The growth of AI/Deep learning and data analytics has created many of the most challenging HPC workloads in recent years. The latest HPC report by Hyperion Research states that iterative simulation workloads and new workloads such as AI and other Big Data jobs would drive the adoption of HPC storage.
To keep up with the growing massive amount of data we are collecting, users need to enhance computation performance at the same time and hence HPC requires equally robust storage to maintain compute performance for faster data in and out as we are heading into the Big Data era now.
Data-intensive HPC is driving new storage requirements and making a change. For the simulation process, it not only requires a large amount of computations running on HPC infrastructures built on a cluster of powerful servers linked together with networking and memory, but also adds in self-service concept data stores. In analytics, in addition to focusing on the storage and access of the data so analytics is performed on a Big Data infrastructure suited for the problem at hand, co-processors and fast networks can also speed up analytics.
HPC storage is a key component for the smooth and efficient running of an HPC cluster. A high density server platform with local storage is an ideal system design to promote data faster. TYAN’s Thunder HX FT83-B7119 features high density local storage within a 4U server platform and is designed to handle expanding workloads involving simulation, analytics, AI, and deep learning. The system is based on dual-socket 2nd Gen Intel® Xeon® Scalable Processors, supporting up to 10 double-width PCIe x16 slots for GPU deployment, up to 3TB DDR4-2933 memory, either twelve 3.5” SATA 6G drive bays or eight SATA plus four NVMe U.2 bays for faster IO, a PCIe x16 slot for a high performance NIC, a BMC with Redfish support, and a 3+1 4800W redundant power supply.
A HPC storage server system like the Thunder HX FT83-B7119 can solve complex and data-intensive tasks to ensure low latency with high bandwidth and application, storage and compute all in the same place. Depending on application needs, the system comes with two types of PCIe routing topology
architecture for different use cases. The balanced configuration has the 10 GPU cards evenly routed to the two CPUs which provides high CPU-to-GPU bandwidth for various parallel workloads such as real-time facial recognition workloads which need access to very large databases; the other configuration has the 10 GPU cards solely routed to the first CPU which is ideal for GPUDirect RDMA applications and AI/Machine Learning/Deep Neural Network workloads. The expansion capability also allows a single-wide x16 card to be installed next to the 10 double-wide PCIe x16 cards, and it supports high speed networking such as 100 Gigabit EDR InfiniBand or Ethernet. This is also a benefit for big-data analytics that can speed up the time it takes to access datasets.
As interest in AI/Deep Learning continues to increase in traditional HPC, the Thunder HX FT83-B7119 is good choice for both types of applications. With Intel Xeon Scalable processors, a large memory foot print, and plenty of storage the FT83-B7119 provides a great base for many HPC and Deep Learning applications.
To learn more about Thunder HX FT83-B7119
To learn more about other Intel Xeon Scalable Processor-based server platforms offered by TYAN, please check out TYAN server solution online exhibition