Emerging data-intensive applications increasingly demand large-scale, cost-effective, and high-performance memory solutions. Samsung's CXL Memory Module-Hybrid (CMM-H) uniquely integrates DDR DRAM and NAND flash storage within a single CXL-attached memory device for higher capacity while keeping still high performance. This paper provides a comprehensive exploration of the CMM-H module, detailing its architectural design, operational workflow, and caching mechanisms. We evaluate the performance of CMM-H through extensive experiments, highlighting its benefits and limitations. Our study contributes insights into hybrid CXL memory architectures and provides valuable guidance for future CXL memory design improvements.

Edge computing is a distributed computing paradigm that collects and processes data at or near the source of data generation. The on-device learning at edge relies on device-to-device wireless communication to facilitate real-time data sharing and collaborative decision-making among multiple devices. This significantly improves the adaptability of the edge computing system to the changing environments. However, as the scale of the edge computing system is getting larger, communication among devices is becoming the bottleneck because of the limited bandwidth of wireless communication leads to large data transfer latency. To reduce the amount of device-to-device data transmission and accelerate on-device learning, in this paper, we propose Residual-INR, a fog computing-based communication-efficient on-device learning framework by utilizing implicit neural representation (INR) to compress images/videos into neural network weights. Residual-INR enhances data transfer efficiency by collecting JPEG images from edge devices, compressing them into INR format at the fog node, and redistributing them for on-device learning. By using a smaller INR for full image encoding and a separate object INR for high-quality object region reconstruction through residual encoding, our technique can reduce the encoding redundancy while maintaining the object quality. Residual-INR is a promising solution for edge on-device learning because it reduces data transmission by up to 5.16 × across a network of 10 edge devices. It also facilitates CPU-free accelerated on-device learning, achieving up to 2.9 × speedup without sacrificing accuracy. Our code is available at: https://github.com/sharc-lab/Residual-INR.

The industry has adopted disaggregated storage systems to provide high-quality services for hyper-scale architectures. This infrastructure enables organizations to access storage resources that can be independently managed, configured, and scaled. It is supported by the recent advances of all-flash arrays and NVMe-over-Fabric protocol, enabling remote access to NVMe devices over different network fabrics. A surge of research has been proposed to mitigate network congestion in traditional remote direct memory access protocol (RDMA). However, NVMe-oF raises new challenges in congestion control for disaggregated storage systems.In this work, we investigate the performance degradation of the read throughput on storage nodes caused by traditional network congestion control mechanisms. We design a storage-side rate control (SRC) to relieve network congestion while avoiding performance degradation on storage nodes. First, we design an I/O throughput control mechanism in the NVMe driver layer to enable throughput control on storage nodes. Second, we construct a throughput prediction model to learn a mapping function between workload characteristics and I/O throughput. Third, we deploy SRC on storage nodes to cooperate with traditional network congestion control on an NVMe-over-RDMA architecture. Finally, we evaluate SRC with varying workloads, SSD configurations, and network topologies. The experimental results show that SRC achieves significant performance improvement.

Deep learning-based recommendation systems are extensively deployed in numerous internet services, including social media, entertainment services, and search engines, to provide users with the most relevant and personalized content. Production scale deep learning models consist of large embedding tables with billions of parameters. DRAM-based recommendation systems incur a high infrastructure cost and limit the size of the deployed models. Recommendation systems based on solid-state drives (SSDs) are a promising alternative for DRAM-based systems. Systems based on SSDs can offer ample storage required for deep learning models with large embedding tables. This paper proposes SmartRec, an inference engine for deep learning-based recommendation systems that utilizes Samsung SmartSSD, an SSD with an on-board FPGA that can process data in-situ. We evaluate SmartRec with state-of-the-art recommendation models from Facebook and compare its performance and energy efficiency to a DRAM-based system on a CPU. We show SmartRec improves the energy efficiency of the recommendation inference task up to 10x in comparison to the baseline CPU implementation. In addition, we propose a novel application-specific caching system for SmartSSDs that allows the kernel on the FPGA to use its DRAM as a cache to minimize high latency SSD accesses. Finally, we demonstrate the scalability of our design by offloading the computation to multiple SmartSSDs to further improve performance.

Due to the so-called curse of dimensionality and increase in the size of databases, there is an ever-increasing demand for computing resources and memory bandwidth when performing the k-nearest neighbors (kNNs) algorithm, resulting in a slow-down to process large datasets. This work presents an OpenCL-based framework for accelerating the kNN algorithm on field-programmable gate arrays (FPGAs) benefiting from the random projection dimensionality reduction. The proposed RPkNN framework includes two compute modules implementing a throughput-optimized hardware architecture based on random projection and the kNN algorithm and a host program facilitating easy integration of the compute modules in the existing applications. RPkNN also utilizes a new buffering scheme tailored to random projection and the kNN algorithm. The proposed architecture enables parallel kNN computations with a single memory channel and takes advantage of the sparsity features of the input data to implement a highly optimized and parallel implementation of random projection. We employ a computation storage device (CSD) to directly access the high-dimensional data on non-volatile memory express (NVMe) solid state drive (SSD) and store and reuse the compressed and low-dimensional data on the FPGA dynamic random access memory (DRAM), hence eliminating data transfers to the host DRAM. We compare RPkNN implemented on the Samsung SmartSSD CSD with the kNN implementation of the scikit-learn library running on an Intel Xeon Gold 6154 CPU. The experimental results show that the proposed RPkNN solution achieves, on average,$26\times $and$46\times $higher performance across different dimensions per a single kNN computation for the SIFT1M and GIST1M databases, respectively. Finally, RPkNN is$1.7\times $faster than the similar FPGA-based reference method.