Related papers: Echo: Efficient Co-Scheduling of Hybrid Online-Off…

The rapid growth of Internet of Things (IoT) devices produces massive, heterogeneous data streams, demanding scalable and efficient scheduling in cloud environments to meet latency, energy, and Quality-of-Service (QoS) requirements.…

Large language models (LLMs) are typically served from clusters of GPUs/NPUs that consist of large number of devices. Unfortunately, communication between these devices incurs significant overhead, increasing the inference latency and cost…

KV caching is a fundamental technique for accelerating Large Language Model (LLM) inference by reusing key-value (KV) pairs from previous queries, but its effectiveness under limited memory is highly sensitive to the eviction policy. The…

Large Language Models have revolutionized natural language processing, yet serving them efficiently in data centers remains challenging due to mixed workloads comprising latency-sensitive (LS) and best-effort (BE) jobs. Existing inference…

Speculative Decoding promises to accelerate the inference of Large Language Models, yet its efficacy often degrades in production-grade serving. Existing evaluations typically overlook the compute-bound nature of high-concurrency regimes,…

High throughput serving of large language models (LLMs) requires batching sufficiently many requests at a time. However, existing systems struggle because the key-value cache (KV cache) memory for each request is huge and grows and shrinks…

Large Language Model (LLM) workloads have distinct prefill and decode phases with different compute and memory requirements which should ideally be accounted for when scheduling input queries across different LLM instances in a cluster.…

In the field of instruction-following large vision-language models (LVLMs), the efficient deployment of these models faces challenges, notably due to the high memory demands of their key-value (KV) caches. Conventional cache management…

Multi-server jobs that request multiple computing resources and hold onto them during their execution dominate modern computing clusters. When allocating the multi-type resources to several co-located multi-server jobs simultaneously in…

While Large Language Models and their underlying Transformer architecture are remarkably efficient, they do not reflect how our brain processes and learns a diversity of cognitive tasks such as language, nor how it leverages working memory.…

Serving systems for Large Language Models (LLMs) improve throughput by processing several requests concurrently. However, multiplexing hardware resources between concurrent requests involves non-trivial scheduling decisions. Practical…

Modern large language models (LLMs) increasingly rely on inference-time planning and external tools to improve reasoning. We benchmark this behavior on two real-world settings: event-centric question answering over graph-structured…

Large language models (LLMs) achieve state-of-the-art accuracy on complex reasoning tasks by generating multiple chain-of-thought (CoT) traces, but using a fixed token budget per query leads to over-computation on easy inputs and…

Test-time reinforcement learning generates multiple candidate answers via repeated rollouts and performs online updates using pseudo-labels constructed by majority voting. To reduce overhead and improve exploration, prior work introduces…

Large Language Models (LLMs) and other large foundation models have achieved noteworthy success, but their size exacerbates existing resource consumption and latency challenges. In particular, the large-scale deployment of these models is…

Volunteer computing is an Internet-based distributed computing system in which volunteers share their extra available resources to manage large-scale tasks. However, computing devices in a Volunteer Computing System (VCS) are highly dynamic…

Recent breakthroughs in large Language Models (LLMs) have enabled various generative tasks on a single model. Real-world services (e.g., OpenAI's ChatGPT [27]) powered by an LLM often concurrently support latency-critical requests for…

Long chain-of-thought reasoning and agentic tool-calling produce traces spanning tens of thousands of tokens, yet Transformer KV caches grow linearly with sequence length, creating a memory bottleneck on commodity hardware. State-space…

Large language models (LLMs) often incorporate multiple text chunks in their inputs to provide the necessary contexts. To speed up the prefill of the long LLM inputs, one can pre-compute the KV cache of a text and re-use the KV cache when…

The demand for stringent interactive quality-of-service has intensified in both mobile edge computing (MEC) and cloud systems, driven by the imperative to improve user experiences. As a result, the processing of computation-intensive tasks…