SNIA Developer Conference September 15-17, 2025 | Santa Clara, CA
KV-CSD: An Ordered, Hardware-Accelerated Key-Value Store For Rapid Data Insertion and Queries
Salon II
Tue Sep 19 | 1:00pm
Abstract
Rapidly increasing data sizes, the high cost of data movement, and the advent of fast, NVMe-over-fabric based flash enclosures have led to the exploration of computation near flash for more efficient and economical storage solutions. Ordered key-value stores, commonly developed as software library code that runs inside application processes such as LevelDB and RocksDB, are one of many storage functions that can potentially benefit from offloaded processing. This is because traditional host managed key-value stores often exhibit long processing delays when their background worker threads cannot sort data as fast as a foreground writer application can write it, due to large amounts of data movement between the host and storage during the sort. Offloading key-value store computation to storage is interesting because it allows those data-intensive background tasks to be deferred and performed asynchronously on storage rather than on a host. This better hides background work latency and prevents it from blocking foreground writes. Offloaded key-value stores are interesting also because the key-value interface itself provides sufficient knowledge of data without requiring external metadata, leaving room for building more types of indexes such as secondary indexes and histograms. In this talk, we present KV-CSD, a research collaboration between SK hynix and Los Alamos National Lab (LANL) that explores the lab's next-generation performance-tier storage designs. A KV-CSD is a key-value based computational storage device consisting of a ZNS NVMe SSD and a System-on-a-Chip (SoC) that implements an ordered key-value store atop the SSD. It supports insertion, deletion, histogram generation, point/range queries over primary keys, and point/range queries over user-defined secondary index keys, a function that is often missed by today’s popular software key-value stores. We show why computational storage in the form of a hardware-accelerated key-value store is particularly interesting to LANL's simulation-based science workflows, how it fits into LANL's overall storage infrastructure designs, and how we implement KV-CSD to address bottlenecks experienced by scientists when high volumes of small data records previously written by a massively parallel simulation are subsequently read for interactive data analytics with potentially very selective queries against multiple data dimensions.
Learning Objectives
- Overview of HPC storage architectures and bottlenecks
- HPC-driven motivations for ordered key-value based computational storage
- A practical implementation of a hardware-accelerated ordered key-value store for large-scale scientific data absorption, process, and analytics
Related Sessions
Media Objectives for Next-Generation Video Games
The consumer video game software industry is on the cusp of its 50th year, and its revenues continue to grow steadily, in recent years outpacing even those of motion pictures. The role of data storage media is uniquely central to the video game industry, as it is ultimately the canvas upon which game developers paint. Despite the convenience of downloadable content via the Internet, strong demand continues for games on physical storage media, especially from major publishers.
Games from major publishers also require more storage than ever, not only for their distribution, but also for logging data about game play, from high score lists to saved session information and sharing playback videos. But the growth in the industry’s distribution media has slowed, from more than 1000 times over three generations of semiconductor ROM cartridges to hardly more than 150 times over five generations of optical discs. So what’s next?
This session reprises a popular 2019 SDC presentation surveying the significant roles and evolving types of digital storage media that enable this interactive digital art form, updated to include the latest generation of video game systems.
OCP Storage Project Update
This is an update on the activities in the OCP Storage Project.
Data-Intensive Inference Done Better: Scaling Models and RAG in Limited Memory with SSD Offload
Enterprises are rushing to adopt AI inference solutions with RAG to solve business problems, but enthusiasm for the technology's potential is outpacing infrastructure readiness. It quickly becomes prohibitively expensive or even impossible to use more complex models and bigger RAG data sets due to the cost of memory. Using open-source software components and high-performance NVMe SSDs, we explore two different but related approaches for solving these challenges and unlocking new levels of scale: offloading model weights to storage using DeepSpeed, and offloading RAG data to storage using DiskANN. By combining these, we can achieve (a) more complex models running on GPUs that it was previously impossible to use, and (b) greater cost efficiency when using large amounts of RAG data. We'll talk through the approach, share benchmarking results, and show a demo of how the solution works in an example use case.
Chiplets, UCIe, Persistent Memory, and Heterogeneous Integration: The Processor Chip of the Future!
Chiplets have become a near-overnight success with today’s rapid-fire data center conversion to AI. But today’s integration of HBM DRAM with multiple SOC chiplets is only the very beginning of a larger trend in which multiple incompatible technologies will adopt heterogeneous integration to connect new memory technologies with advanced logic chips to provide both significant energy savings and vastly-improved performance at a reduced price point. In this presentation analysts Tom Coughlin and Jim Handy will explain how memory technologies like MRAM, ReRAM, FRAM, and even PCM will eventually displace the DRAM HBM stacks used with xPUs, on-chip NOR flash and SRAM, and even NAND flash in many applications. They will explain how DRAM’s refresh mechanism and NAND and NOR flash’s energy-hogging writes will give way to much cooler memories that will be easier to integrate within the processor’s package, how processor die sizes will dramatically shrink through the use of new memory technologies to replace on-chip NOR and SRAM, and how the UCIe interface will allow these memories to compete to bring down overall costs. They will also show how the approach will not only reduce the purchase price per teraflop, but also how the energy costs per teraflop will also improve.
