NAND Flash

The importance of solid state storage continues to grow for data center operations. Faster than hard drives and cheaper than DRAM, flash memory provides a valuable middle ground. As NAND memory progresses, producers have stopped looking at closer node topologies for planar NAND in favor of 3D NAND arrays to drive solid state drive (SSD) dollars/GB closer to hard drives. To further their appeal, storage controllers can improve upon flash memory units by increasing their efficiency and life expectancy. To learn more about these solutions, visit our SSD for datacenter site.

Storage Class Memories

A new class of memory solutions that promises to fill the gap between DRAM and NAND. The tradeoff of memories are always between speed, cost/bit, and write endurance. With this new class and capabilities, performance will improve for a multitude of tasks. These tasks can be memory area networking, memory expansion, write acceleration, and write commitment  are all possible. These memories also can reduce data loss due to power outages, super cap failures just to name a few. For more information, please visit our memory and storage solutions. 

For high-performance computing, FPGAs offer the low-latency offloading necessary to accelerate functions, such as design modeling, oil and gas search, nuclear power simulations, and other functions. FPGAs support in-line processing to offload CPU workloads by reducing complex bottlenecks. Additionally, FPGAs provide the ability for implementations of secure hash algorithm (SHA), de-dupe capabilities, erasure coding, and various forms of compression such as gzip. This in-line processing benefits system architecture dually by freeing up limited processor memory while lowering the computational load placed on the processor. By doing so, the FPGA can reduce power consumption and be in the optimized location for maximum performance in the data center.

Solid state drives (SSD) offer a significant performance advantage over hard drives; the bandwidth improvements simply cannot be matched. However, over SAS or SATA busses this potential performance edge is severely limited. With non-volatile memory express (NVMe) over PCIe*, SSD bandwidths reach nearly 3X or 4X better performance than over SAS/SATA respectively.

Via FPGA PCIe* plug-in cards many bottlenecks resulting from SAS and SATA systems can be alleviated to improve storage networks in enterprise workloads. NVMe has since expanded to include capabilities over fabric. Intel offers RDMA over converged Ethernet (RoCE) to improve storage structures including native latencies for remote SSDs.

Software-defined storage (SDS) has become a rapidly growing industry trend. Remote storage systems are managed with a software-defined network responsible for replication and backup among other tasks to virtualize the storage network. However, this network can be improved via hardware with FPGAs to attain near native SSD latency, throughput, and IOPS despite being remote storage systems. This FPGA solution utilizes NVMe over RoCE, where Intel® FPGAs can act as both the host interface and the storage controller necessary in the SDS network.

In RoCE implementations, storage area networks offload server network responsibilities to the FPGA relieving both CPU workloads and memory. With RoCE, server-to-server data storage transfer does not require the CPU and the process can be implemented with very low network latencies. In doing so, CPU memory buffers can be freed up for more pressing processing needs, improving the overall functionality of the servers.