Related papers: One Pool, Two Caches: Adaptive HBM Partitioning fo…

LLM inference powers latency-critical production services nowadays. The bursty nature of inference traffic results in over-provisioning, which in turn leads to resource underutilization. While online-offline colocation promises to utilize…

Stream analytics have an insatiable demand for memory and performance. Emerging hybrid memories combine commodity DDR4 DRAM with 3D-stacked High Bandwidth Memory (HBM) DRAM to meet such demands. However, achieving this promise is…

Inference on large-language models (LLMs) is constrained by GPU memory capacity. A sudden increase in the number of inference requests to a cloud-hosted LLM can deplete GPU memory, leading to contention between multiple prompts for limited…

Traditional memory management suffers from metadata overhead, architectural complexity, and stability degradation, problems intensified in cloud environments. Existing software/hardware optimizations are insufficient for cloud computing's…

The widespread adoption of LLMs has driven an exponential rise in their deployment, imposing substantial demands on inference clusters. These clusters must handle numerous concurrent queries for different LLM downstream tasks. To handle…

LLM decoding is bottlenecked for large batches and long contexts by loading the key-value (KV) cache from high-bandwidth memory, which inflates per-token latency, while the sequential nature of decoding limits parallelism. We analyze the…

Large Language Models (LLMs) in agentic workflows combine multi-step reasoning, heterogeneous tool use, and collaboration across multiple specialized agents. Existing LLM serving engines optimize individual calls in isolation, while…

Modern day applications have grown in size and require more computational power. The rise of machine learning and AI increased the need for parallel computation, which has increased the need for GPGPUs. With the increasing demand for…

Large language model (LLM) inference serving systems are essential to various LLM-based applications. As demand for LLM services continues to grow, scaling these systems to handle high request rates while meeting latency Service-Level…

Distributed prefix caching has become a core technique for efficient LLM serving. However, for long-context requests with high cache hit ratios, retrieving reusable KVCache blocks from remote servers has emerged as a new performance…

Deep learning recommendation models (DLRMs) have been widely applied in Internet companies. The embedding tables of DLRMs are too large to fit on GPU memory entirely. We propose a GPU-based software cache approaches to dynamically manage…

This paper presents Block, a distributed scheduling framework designed to optimize load balancing and auto-provisioning across instances in large language model serving frameworks by leveraging contextual information from incoming requests.…

This paper introduces Helix, a distributed system for high-throughput, low-latency large language model (LLM) serving in heterogeneous GPU clusters. The key idea behind Helix is to formulate inference computation of LLMs over heterogeneous…

Current HPC systems provide memory resources that are statically configured and tightly coupled with compute nodes. However, workloads on HPC systems are evolving. Diverse workloads lead to a need for configurable memory resources to…

Scaling quantum computing beyond a single device requires networking many quantum processing units (QPUs) into a coherent quantum-HPC system. We propose the Modular Entanglement Hub (ModEn-Hub) architecture: a hub-and-spoke photonic…

Deploying large language models (LLMs) for online inference is often constrained by limited GPU memory, particularly due to the growing KV cache during auto-regressive decoding. Hybrid GPU-CPU execution has emerged as a promising solution…

Personalized recommendation is a ubiquitous application on the internet, with many industries and hyperscalers extensively leveraging Deep Learning Recommendation Models (DLRMs) for their personalization needs (like ad serving or movie…

Traditional GPU hash tables preserve every inserted key -- a dictionary assumption that wastes scarce High Bandwidth Memory (HBM) when embedding tables routinely exceed single-GPU capacity. We challenge this assumption with cache semantics,…

Latency-sensitive inference on shared A100 clusters often suffers noisy-neighbor interference on the PCIe fabric, inflating tail latency and SLO violations. We present a fabric-agnostic, VM-deployable host-level controller that combines…

Modern LLM serving is increasingly serverless in shape: large model catalogs, long-tail invocations, and multi-tenant demand. Existing GPU serving systems face a tradeoff: dedicated-GPU allocation wastes scarce HBM under sparse traffic,…