Related papers: CXL Memory as Persistent Memory for Disaggregated …
Large Language Models (LLMs) are becoming increasingly data-intensive due to growing model sizes, and they are becoming memory-bound as the context length and, consequently, the key-value (KV) cache size increase. Inference, particularly…
Upcoming CXL-based disaggregated memory devices feature special purpose units to offload compute to near-memory. In this paper, we explore opportunities for offloading compute to general purpose cores on CXL memory devices, thereby enabling…
On heterogeneous memory (HM) where fast memory (i.e., CPU-attached DRAM) and slow memory (e.g., remote NUMA memory, RDMA-connected memory, Persistent Memory (PM)) coexist, optimizing the placement of tree-structure indexes (e.g., B+tree) is…
Memory disaggregation via CXL enables multi-host resource sharing. However, existing CXL sharing mechanisms enforce coarse-grained, host-level permissions only, leaving isolation to the operating system. Today, virtual memory enables…
Emerging Persistent Memory technologies (also PM, Non-Volatile DIMMs, Storage Class Memory or SCM) hold tremendous promise for accelerating popular data-management applications like in-memory databases. However, programmers now need to deal…
Despite the promise of alleviating the main memory bottleneck, and the existence of commercial hardware implementations, techniques for Near-Data Processing have seen relatively little real-world deployment. The idea has received renewed…
The configurable building blocks of current FPGAs -- Logic blocks (LBs), Digital Signal Processing (DSP) slices, and Block RAMs (BRAMs) -- make them efficient hardware accelerators for the rapid-changing world of Deep Learning (DL).…
Memory disaggregation has attracted great attention recently because of its benefits in efficient memory utilization and ease of management. So far, memory disaggregation research has all taken one of two approaches: building/emulating…
Contrast maximization (CMAX) is a direct geometric framework for event-based motion estimation, but its iterative warp-and-accumulate pipeline incurs input-dependent computation and frequent memory accesses, challenging real-time, low-power…
CXL (Compute Express Link) enables multiple hosts to share byte-addressable memory with hardware cache coherence, but no existing filesystem exploits this for lock-free multi-host coordination. We present DaxFS, a Linux filesystem for CXL…
Public cloud providers seek to meet stringent performance requirements and low hardware cost. A key driver of performance and cost is main memory. Memory pooling promises to improve DRAM utilization and thereby reduce costs. However,…
The advent of non-volatile memory (NVM) technologies like PCM, STT, memristors and Fe-RAM is believed to enhance the system performance by getting rid of the traditional memory hierarchy by reducing the gap between memory and storage. This…
High-performance Host processors can integrate Processing-In-Memory (PIM) devices, which can accelerate memory-intensive kernels of Machine Learning (ML) models, including Large Language Models (LLMs), by leveraging the large memory…
Parallel programming remains a daunting challenge, from the struggle to express a parallel algorithm without cluttering the underlying synchronous logic, to describing which devices to employ in a calculation, to correctness. Over the…
This paper proposes ScalePool, a novel cluster architecture designed to interconnect numerous accelerators using unified hardware interconnects rather than traditional long-distance networking. ScalePool integrates Accelerator-Centric Links…
Computational memory (CM) is a promising approach for accelerating inference on neural networks (NN) by using enhanced memories that, in addition to storing data, allow computations on them. One of the main challenges of this approach is…
Modern AI workloads such as large language models (LLMs) and retrieval-augmented generation (RAG) impose severe demands on memory, communication bandwidth, and resource flexibility. Traditional GPU-centric architectures struggle to scale…
The conventional approach of moving data to the CPU for computation has become a significant performance bottleneck for emerging scale-out data-intensive applications due to their limited data reuse. At the same time, the advancement in 3D…
Compute and memory are tightly coupled within each server in traditional datacenters. Large-scale datacenter operators have identified this coupling as a root cause behind fleet-wide resource underutilization and increasing Total Cost of…
The global scarcity of GPUs necessitates more sophisticated strategies for Deep Learning jobs in shared cluster environments. Accurate estimation of how much GPU memory a job will require is fundamental to enabling advanced scheduling and…