Samsung and Xilinx partner up to create the world’s smartest SSD

With their SmartSSD, Samsung is combining their NAND storage with a Xilinx FPGA to accelerate a variety of storage applications, moving database management and processing away from CPUs and placing it directly onto storage mediums. This allows Samsung’s SmartSSD to deliver significantly faster storage speeds, higher data compression ratios and allows software developers to create new hardware-accelerated storage solutions. 

Samsung’s current SmartSSD is a 4TB drive, which can be used to store up to 12TB of data through accelerated transparent compression. This compression can be used to deliver up to 10x more storage capacity in some use cases.  

With SmartSSD storage, some applications can be accelerated to offer users significant increases to system performance. For big data analytics, Samsung’s  SmartSSD CSD with Bigstream’s hyper-acceleration layer allows customers to increase time-to-insight by up to 10x on the Apache Spark unified analytics engine for big data. For search-in-storage operations, the Lewis Rhodes Labs (LRL) applies novel neuromorphic processing with the SmartSSD CSD to accelerate searches by up to 100x, allowing petabytes to be searched in minutes. 

Thank to Xilinx’ onboard SSD processing, storage overhead is taken away from the user’s processor, allowing that performance to be reallocated to other tasks. On the Xbox Series X/S and PlayStation 5, both console manufacturers have already taken similar steps by adding dedicated decompression units to their systems, removing some of the storage overhead from their processors. Samsung and Xilinx’ solution does the same thing on a broader scale, offering users more than simple decompression acceleration.