A novel monolithic 3D-stackable (M3D) DRAM-based processing-near-memory (PNM) architecture that eliminates memory bottlenecks using a Cu-Cu hybrid-bonded logic die integrated with an M3D DRAM die is proposed, and hybrid binary--stochastic floating-point multipliers are employed.
Abstract
Normalizing Flows (NFs) have shown strong results in generative modeling, recently reaching state-of-the-art image generation performance. However, their repetitive, sequential operations limit the computational and memory efficiency achievable with GPUs and TPUs. We propose M3D-FLOW, a novel monolithic 3D-stackable (M3D) DRAM-based processing-near-memory (PNM) architecture that eliminates memory bottlenecks using a Cu-Cu hybrid-bonded logic die integrated with an M3D DRAM die. The M3D DRAM provides dense bank integration and a tiered organization with variable access latencies that M3D-FLOW exploits for the efficient processing of different NF workloads. To maximize compute density while minimizing area, energy, and thermal overhead, M3D-FLOW employs hybrid binary--stochastic floating-point multipliers, performing mantissa multiplication in the stochastic domain and exponent operations in the binary domain. A pipelined execution strategy maps sequential NF stages across dynamically partitioned PE groups, and together with the stochastic hardware, enables M3D-FLOW to achieve at least $6.6\times$ higher performance and $3\times$ better energy efficiency than prior PNM accelerators, with only a 2.2% bits-per-dimension (BPD) increase over bfloat16 and 4.1% over FP32. An analysis of stochastic hardware variants further shows that shorter stochastic bitstreams reduce execution time by up to $3.3\times$ at the cost of higher approximation error, providing a concrete design guide for latency--accuracy trade-offs.
Processing-in-Memory (PIM) on monolithic 3D (M3D) DRAM is a promising answer to the memory wall for data-intensive dynamic programming (DP), yet extracting its performance today demands hand-written kernels: the programmer must pick a tile size, place data across non-uniform-latency memory tiers, partition work between...
Mahbod Afarin, Tsung-Han Lu, T. Rosing· 0 citations
4D Gaussian Splatting (4DGS) enables high-quality real-time novel-view synthesis for dynamic scenes, supporting free-viewpoint video, mobile XR, and autonomous systems at the edge. Methods using explicit 4D distributions, keyframe interpolation, or anchor binding share two GPU data-movement bottlenecks. Temporal filter...
Zi-Rui Ma, Zhi-Hua Fan, Hai-Bin Wu et al.· 0 citations
Computing-in-Memory using DRAM (CIMD) has demonstrated substantial energy and throughput gains for memory-bound workloads consisting of bulk-bitwise operations, by performing computation directly within DRAM subarrays. Realizing CIMD, however, requires a redesign of the memory controller and careful mapping of operands...
Alexander Kusnezoff, J. P. C. de Lima, J. Castrillón et al.· 0 citations
Mixture-of-Experts (MoE) have become a mainstream way to scale Large Language Models (LLMs), because each token activates only a small subset of experts, reducing inference cost. However, MoE decode remains highly memory-bound and exhibits strongly skewed, time-varying expert loads. 3D-stacked DRAM near-memory processi...
Yu-Dong Pan, Lian Liu, Shi-Xin Zhao et al.· 0 citations
Rapid advancements in large-scale deep neural networks (DNNs) have increased the demand for substantial data storage and memory bandwidth, resulting in the well-known data movement bottleneck in traditional von Neumann-based architectures. Processing-in-memory (PIM) technology, which integrates computation within memor...
Inseong Hwang, Jihoon Jang, C. Park et al.· ACM Transactions on Reconfig...· 0 citations
On-demand model loading dominates response latency for interactive on-device AI, yet unified memory architectures (UMAs), now common on edge devices, should make it cheap by placing weights in memory the GPU already sees. However, frameworks still use the loading flow built for discrete GPUs, leaving significant ineffi...
Zi-Meng Wang, Jin-Rong Yang, Rong Chen· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.