Relay Buffer Independent Communication over Pooled HBM for Efficient MoE Inference on Ascend

• URL: http://arxiv.org/abs/2605.06055v2
• Date: Fri, 08 May 2026 01:46:43 GMT
• Title: Relay Buffer Independent Communication over Pooled HBM for Efficient MoE Inference on Ascend
• Abstract summary: Mixture-of-Experts (MoE) inference requires large-scale token exchange across devices.<n>This report presents a relay-buffer-free communication design for MoE inference acceleration on Ascend systems.
• Score: 14.495939227785074
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Mixture-of-Experts (MoE) inference requires large-scale token exchange across devices, making dispatch and combine major bottlenecks in both prefill and decode. Beyond network transfer, routing-driven layout transformation, temporary relay, and output restoration can add substantial overhead. Existing MoE communication paths are often buffer-centric, using explicit inter-process relay and reordering buffers around collective transfer. This report presents a relay-buffer-free communication design for MoE inference acceleration on Ascend systems. The design reorganizes dispatch and combine around direct placement into destination expert windows and direct reading from remote expert windows. Built on globally pooled high-bandwidth memory and symmetric-memory allocation, it removes most intermediate relay and reordering buffers while retaining only lightweight control state, including counts, offsets, and synchronization metadata. We instantiate the design as two schedules for the main phases of MoE inference: a prefill schedule with richer planning state for throughput-oriented execution, and a compact decode schedule for latency-sensitive execution. Experiments on Ascend-based MoE workloads show reduced dispatch and combine latency in both settings. At the serving level, the implementation improves time to first token (TTFT), preserves competitive time per output token (TPOT), and enlarges the feasible scheduling space under practical latency constraints. These results indicate that, on platforms with globally addressable device memory, reducing intermediate buffering and output restoration around expert execution is an effective direction for accelerating MoE inference.

Related papers

• Accelerating Unified Multimodal Models with Core-Expansion Routing and Unified Computation Scheduling [61.28599393707943]
  Unified multimodal models jointly support understanding and generation, but incur substantial redundant computation across tokens, layers, and generation timesteps.<n>We propose a task-shared core scorer and progress-conditioned generation expansions, optimized through generation decomposition and cross-task core alignment.<n> Experiments on two representative UMM architectures demonstrate consistent quality--efficiency improvements across both tasks, retaining 98.03% of dense understanding performance with a 1.93times$ end-to-end inference speedup.
  arXiv  Detail & Related papers  (2026-08-29T14:27:55Z)
• DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation [69.8840101036735]
  Speculative decoding accelerates Large Language Model (LLM) inference.<n>Recent parallel drafters efficiently propose long token sequences in a single forward pass, but suffer from rapid acceptance decay due to a lack of inter-token dependencies.<n>We introduce DSpark, a speculative decoding framework that unifies high- throughput parallel generation with adaptive, load-aware verification.
  arXiv  Detail & Related papers  (2026-07-06T14:28:06Z)
• An Efficient vLLM-Based Inference Pipeline for Unified Audio Understanding and Generation [63.50393485685279]
  We present a vLLM-based inference pipeline for unified speech understanding and generation.<n>We overcome the shared intuition-Free Guidance (CFG) halves throughput.<n>By co-scheduling paired conditional and unconditional requests within a continuous batch, our CFG implementation sustains 80% of non-CFG throughput.
  arXiv  Detail & Related papers  (2026-07-02T12:55:11Z)
• CoMem: Context Management with A Decoupled Long-Context Model [68.6597154502361]
  CoMem is a framework that decouples memory management from the primary agent workflow.<n>We show that CoMem provides 1.4x latency improvements upon vanilla long-context solutions.
  arXiv  Detail & Related papers  (2026-05-29T04:59:32Z)
• Rethinking Memory as Continuously Evolving Connectivity [63.58181883607843]
  Existing memory-augmented LLM agents treat memory as a static repository with pre-defined representations and fixed retrieval pipelines.<n>We propose FluxMem, a connectivity-evolving memory framework that models memory as a heterogeneous graph and progressively refines its topology through three stages.<n>Across three fundamentally distinct benchmarks including LoCoMo, Mind2Web, and GAIA, FluxMem achieves consistent state-of-the-art performance, demonstrating strong adaptation and generalization in complex agentic environments.
  arXiv  Detail & Related papers  (2026-05-27T17:35:34Z)
• Stream2LLM: Overlap Context Streaming and Prefill for Reduced Time-to-First-Token (TTFT) [2.8350851063316873]
  Streaming context incrementally--overlapping retrieval with inference--can mitigate this latency.<n>We present Stream2LLM, a streaming-aware LLM serving system for concurrent prefill-decode disaggregated deployments.
  arXiv  Detail & Related papers  (2026-03-29T06:49:12Z)
• TS-Memory: Plug-and-Play Memory for Time Series Foundation Models [63.21390142212087]
  Time Series Foundation Models (TSFMs) achieve strong zero-shot forecasting through large-scale pre-training.<n>Existing solutions face a trade-off: Parametric Adaptation can cause catastrophic forgetting, while Non-Parametric Retrieval improves forecasts but incurs high latency due to datastore search.<n>We propose Parametric Memory Distillation and implement it as TS-Memory, a lightweight memory adapter that augments frozen TSFMs.
  arXiv  Detail & Related papers  (2026-02-12T04:16:19Z)
• ZipMoE: Efficient On-Device MoE Serving via Lossless Compression and Cache-Affinity Scheduling [56.88966608455977]
  ZipMoE exploits the synergy between the hardware properties of edge devices and the statistical redundancy inherent to MoE parameters.<n>ZipMoE achieves up to $72.77%$ inference latency reduction and up to $6.76times$ higher throughput than the state-of-the-art systems.
  arXiv  Detail & Related papers  (2026-01-29T02:51:59Z)
• Fast SAM2 with Text-Driven Token Pruning [52.8350457627401]
  Segment Anything Model 2 (SAM2), a vision computation model has significantly advanced in prompt-driven video object segmentation.<n>SAM2 pipelines propagate all visual tokens produced by the image encoder through downstream temporal reasoning modules, regardless of their relevance to the target object.<n>We introduce a text-guided token pruning framework that improves inference efficiency by selectively reducing token density prior to temporal propagation.
  arXiv  Detail & Related papers  (2025-12-24T18:59:05Z)
• Staggered Batch Scheduling: Co-optimizing Time-to-First-Token and Throughput for High-Efficiency LLM Inference [17.27010833526918]
  Staggered Batch Scheduling (SBS) buffers requests to form optimal execution batches.<n> Load-Aware Global Allocation strategy balances computational load across DP units for both Prefill and Decode phases.<n>Our system reduces TTFT by 30%-40% and improves throughput by 15%-20% compared to state-of-the-art immediate scheduling baselines.
  arXiv  Detail & Related papers  (2025-12-18T03:45:05Z)
• Hardware Software Optimizations for Fast Model Recovery on Reconfigurable Architectures [4.058950730052848]
  We present MERINDA, an FPGA-accelerated MR framework that restructures computation as a streaming dataflow pipeline.<n>On representative MR workloads, MERINDA delivers up to 6.3x fewer cycles than an FPGA-based LTC baseline.
  arXiv  Detail & Related papers  (2025-12-05T19:38:34Z)
• DuetServe: Harmonizing Prefill and Decode for LLM Serving via Adaptive GPU Multiplexing [15.376910065679994]
  DuetServe is a unified LLM serving framework that achieves disaggregation-level isolation within a single GPU.<n>DuetServe improves total throughput by up to 1.3x while maintaining low generation latency compared to state-of-the-art frameworks.
  arXiv  Detail & Related papers  (2025-11-06T20:18:34Z)
• Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognition [95.54406667705999]
  Pangu Embedded is an efficient Large Language Model (LLM) reasoner developed on Ascend Neural Processing Units (NPUs)<n>It addresses the significant computational costs and inference latency challenges prevalent in existing reasoning-optimized LLMs.<n>It delivers rapid responses and state-of-the-art reasoning quality within a single, unified model architecture.
  arXiv  Detail & Related papers  (2025-05-28T14:03:02Z)
• Task-Oriented Feature Compression for Multimodal Understanding via Device-Edge Co-Inference [54.53508601749513]
  We propose a task-oriented feature compression (TOFC) method for multimodal understanding in a device-edge co-inference framework.<n>To enhance compression efficiency, multiple entropy models are adaptively selected based on the characteristics of the visual features.<n>Results show that TOFC achieves up to 52% reduction in data transmission overhead and 63% reduction in system latency.
  arXiv  Detail & Related papers  (2025-03-17T08:37:22Z)
• Fast Distributed Inference Serving for Large Language Models [12.703624317418237]
  We present FastServe, a distributed inference serving system for large language models (LLMs) FastServe exploits the autoregressive pattern of LLM inference to enable preemption at the granularity of each output token. We build a system prototype of FastServe and experimental results show that compared to the state-of-the-art solution vLLM, FastServe improves the throughput by up to 31.4x and 17.9x under the same average and tail latency requirements, respectively.
  arXiv  Detail & Related papers  (2023-05-10T06:17:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.