UC San Diego

Existing storage technologies (e.g., disks and flash) are failing to cope with the processor and main memory speed and are limiting the overall performance of many large scale I/O or data-intensive applications. Emerging fast byte-addressable non-volatile memory (NVM) technologies, such as phase-change memory (PCM), spin-transfer torque memory (STTM) and memristor are very promising and are approaching DRAM-like performance with lower power consumption and higher density as process technology scales. These new memories are narrowing down the performance gap between the storage and the main memory and are putting forward challenging problems on existing SSD architecture, I/O interface (e.g, SATA, PCIe) and software. This dissertation addresses those challenges and presents a novel SSD architecture called XSSD. XSSD offloads computation in storage to exploit fast NVMs and reduce the redundant data traffic across the I/O bus. XSSD offers a flexible RPC-based programming framework that developers can use for application development on SSD without dealing with the complication of the underlying architecture and communication management. We have built a prototype of XSSD on the BEE3 FPGA prototyping system. We implement various data-intensive applications and achieve speedup and energy efficiency of 1.5-8.9x and 1.7-10.27x respectively. This dissertation also compares XSSD with previous work on intelligent storage and intelligent memory. The existing ecosystem and these new enabling technologies make this system more viable than earlier ones