arXiv subjects

Minlan Yu

Publications and source records attributed to Minlan Yu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Intra-request branch orchestration for efficient LLM reasoning

Large Language Models (LLMs) increasingly rely on inference-time reasoning algorithms such as chain-of-thought and multi-branch reasoning to improve accuracy on complex tasks. These methods, however, substantially increase token usage and per-request latency. Prior work has largely focused on reducing token usage, often at the expense of accuracy, while overlooking other latency factors. We present DUCHESS, an LLM serving system that reduces cost and latency without sacrificing accuracy through intra-request branch orchestration guided by predictions. DUCHESS employs a lightweight linear probing model over LLM layer activations to estimate branch correctness, and its orchestration policy decides whether to terminate, duplicate, or continue a branch. When handling multiple requests, DUCHESS further reduces latency by prioritizing easier reasoning tasks when complexity can be estimated from the prompt. Experiments on three reasoning benchmarks show that DUCHESS consistently improves the token-accuracy Pareto frontier, reducing token usage by 42-63% at matched accuracy compared to self-consistency. In serving with vLLM, DUCHESS reduces mean, median, and tail latencies by 57-81%, 58-85%, and 52-84% with First-Come-First-Served scheduling, and achieves additional gains under difficulty-aware scheduling at higher request rates.

cs.LG↗

arXiv2025

From Score Distributions to Balance: Plug-and-Play Mixture-of-Experts Routing

Mixture-of-Experts (MoE) models can scale parameter capacity by routing each token to a subset of experts through a learned gate function. While conditional routing reduces training costs, it shifts the burden on inference memory: expert parameters and activations consume memory, limiting the number of experts per device. As tokens are routed, some experts become overloaded while others are underutilized. Because experts are mapped to GPUs, this imbalance translates directly into degraded system performance in terms of latency, throughput, and cost. We present LASER, a plug-and-play, inference-time routing algorithm that balances load while preserving accuracy. LASER adapts to the shape of the gate's score distribution. When scores provide a clear preference, it routes to the strongest experts; when scores are more uniform, it broadens the set of viable experts and routes to the least-loaded among them. Because LASER relies only on gate scores from a trained model, it integrates directly into existing MoE inference pipelines without retraining or finetuning. We evaluate LASER on Mixtral-8x7B and DeepSeek-MoE-16b-chat across four datasets (ARC-Easy, ARC-Challenge, MMLU, and GSM8K). LASER improves load balancing, translating into lower latency and higher throughput, while keeping the accuracy changes negligible.

cs.LG↗

arXiv2025

An LLM-based Agentic Framework for Accessible Network Control

Traditional approaches to network management have been accessible only to a handful of highly-trained network operators with significant expert knowledge. This creates barriers for lay users to easily manage their networks without resorting to experts. With recent development of powerful large language models (LLMs) for language comprehension, we design a system to make network management accessible to a broader audience of non-experts by allowing users to converse with networks in natural language. To effectively leverage advancements in LLMs, we propose an agentic framework that uses an intermediate representation to streamline configuration across diverse vendor equipment, retrieves the network state from memory in real-time, and provides an interface for external feedback. We also conduct pilot studies to collect real user data of natural language utterances for network control, and present a visualization interface to facilitate dialogue-driven user interaction and enable large-scale data collection for future development. Preliminary experiments validate the effectiveness of our proposed system components with LLM integration on both synthetic and real user utterances. Through our data collection and visualization efforts, we pave the way for more effective use of LLMs and democratize network control for everyday users.

cs.NI↗

arXiv2025

ShadowServe: Interference-Free KV Cache Fetching for Distributed Prefix Caching

Distributed prefix caching accelerates long-context LLM serving by reusing KV cache entries for common context prefixes. However, KV cache fetches can become a bottleneck when network bandwidth is limited. Compression mitigates the bandwidth issue, but can degrade overall performance when decompression interferes with model computation. We present ShadowServe, the first SmartNIC-accelerated, interference-free prefix caching system for LLM serving. ShadowServe separates a control plane on the host and a data plane fully offloaded to the SmartNIC, which eliminates interference to both host GPU and CPU. To overcome the SmartNIC's limited compute and memory resources, we design a chunked pipeline that parallelizes data plane operations across the SmartNIC's compute resources, and a minimal-copy memory management scheme that reduces memory pressure on the SmartNIC. Compared to state-of-the-art solutions, ShadowServe achieves up to 2.2x lower loaded time-per-output-token (TPOT), and reduces time-to-first-token (TTFT) by up to 1.38x in low-bandwidth scenarios (<= 20 Gbps), translating to up to 1.35x higher throughput.

cs.DC↗

arXiv2025

Mycroft: Tracing Dependencies in Collective Communication Towards Reliable LLM Training

Reliability is essential for ensuring efficiency in LLM training. However, many real-world reliability issues remain difficult to resolve, resulting in wasted resources and degraded model performance. Unfortunately, today's collective communication libraries operate as black boxes, hiding critical information needed for effective root cause analysis. We propose Mycroft, a lightweight distributed tracing and root cause analysis system designed to address previously hidden reliability issues in collective communication. Mycroft's key idea is to trace collective communication states and leverage internal control and data dependencies to resolve reliability problems in LLM training. Mycroft has been deployed at ByteDance for over six months to debug collective communication related issues at runtime. It detected anomalies within 15 seconds in 90% of cases and identified the root cause within 20 seconds in 60% of cases. We also conducted extensive fault injection experiments to demonstrate Mycroft's capability and efficiency.

cs.DC↗

arXiv2025

Decouple and Decompose: Scaling Resource Allocation with DeDe

Efficient resource allocation is essential in cloud systems to facilitate resource sharing among tenants. However, the growing scale of these optimization problems have outpaced commercial solvers commonly employed in production. To accelerate resource allocation, prior approaches either customize solutions for narrow domains or impose workload-specific assumptions. In this work, we revisit real-world resource allocation problems and uncover a common underlying structure: the vast majority of these problems are inherently separable, i.e., they optimize the aggregate utility of individual resource and demand allocations, under separate constraints for each resource and each demand. Building on this observation, we develop DeDe, a scalable and theoretically rooted optimization framework for large-scale resource allocation. At the core of DeDe is a decouple-and-decompose approach: it decouples entangled resource and demand constraints and thereby decomposes the overall optimization into alternating per-resource and per-demand subproblems that can be solved efficiently and in parallel. We have implemented and released DeDe as a Python package with a familiar modeling interface. Our experiments on three representative resource allocation tasks -- cluster scheduling, traffic engineering, and load balancing -- demonstrate that DeDe delivers significant speedups while generating higher-quality allocations.

cs.DC↗

arXiv2025

Optimus: Accelerating Large-Scale Multi-Modal LLM Training by Bubble Exploitation

Multimodal large language models (MLLMs) have extended the success of large language models (LLMs) to multiple data types, such as image, text and audio, achieving significant performance in various domains, including multimodal translation, visual question answering and content generation. Nonetheless, existing systems are inefficient to train MLLMs due to substantial GPU bubbles caused by the heterogeneous modality models and complex data dependencies in 3D parallelism. This paper proposes Optimus, a distributed MLLM training system that reduces end-to-end MLLM training time. Optimus is based on our principled analysis that scheduling the encoder computation within the LLM bubbles can reduce bubbles in MLLM training. To make scheduling encoder computation possible for all GPUs, Optimus searches the separate parallel plans for encoder and LLM, and adopts a bubble scheduling algorithm to enable exploiting LLM bubbles without breaking the original data dependencies in the MLLM model architecture. We further decompose encoder layer computation into a series of kernels, and analyze the common bubble pattern of 3D parallelism to carefully optimize the sub-millisecond bubble scheduling, minimizing the overall training time. Our experiments in a production cluster show that Optimus accelerates MLLM training by 20.5%-21.3% with ViT-22B and GPT-175B model over 3072 GPUs compared to baselines.

cs.CL↗

arXiv2025

Towards Easy and Realistic Network Infrastructure Testing for Large-scale Machine Learning

This paper lays the foundation for Genie, a testing framework that captures the impact of real hardware network behavior on ML workload performance, without requiring expensive GPUs. Genie uses CPU-initiated traffic over a hardware testbed to emulate GPU to GPU communication, and adapts the ASTRA-sim simulator to model interaction between the network and the ML workload.

cs.NI↗

arXiv2025

Minder: Faulty Machine Detection for Large-scale Distributed Model Training

Large-scale distributed model training requires simultaneous training on up to thousands of machines. Faulty machine detection is critical when an unexpected fault occurs in a machine. From our experience, a training task can encounter two faults per day on average, possibly leading to a halt for hours. To address the drawbacks of the time-consuming and labor-intensive manual scrutiny, we propose Minder, an automatic faulty machine detector for distributed training tasks. The key idea of Minder is to automatically and efficiently detect faulty distinctive monitoring metric patterns, which could last for a period before the entire training task comes to a halt. Minder has been deployed in our production environment for over one year, monitoring daily distributed training tasks where each involves up to thousands of machines. In our real-world fault detection scenarios, Minder can accurately and efficiently react to faults within 3.6 seconds on average, with a precision of 0.904 and F1-score of 0.893.

cs.DC↗

arXiv2025

HACK: Homomorphic Acceleration via Compression of the Key-Value Cache for Disaggregated LLM Inference

Disaggregated Large Language Model (LLM) inference has gained popularity as it separates the computation-intensive prefill stage from the memory-intensive decode stage, avoiding the prefill-decode interference and improving resource utilization. However, transmitting Key-Value (KV) data between the two stages can be a bottleneck, especially for long prompts. Additionally, the computation time overhead for prefill and decode is key for optimizing Job Completion Time (JCT), and KV data size can become prohibitive for long prompts and sequences. Existing KV quantization methods can alleviate the transmission bottleneck and reduce memory requirements, but they introduce significant dequantization overhead, exacerbating the computation time. We propose Homomorphic Acceleration via Compression of the KV cache (HACK) for disaggregated LLM inference. HACK eliminates the heavy KV dequantization step, and directly performs computations on quantized KV data to approximate and reduce the cost of the expensive matrix-multiplication step. Extensive trace-driven experiments show that HACK reduces JCT by up to 70.9% compared to disaggregated LLM inference baseline and by up to 52.3% compared to state-of-the-art KV quantization methods.

cs.DC↗

arXiv2025

EdgeSight: Enabling Modeless and Cost-Efficient Inference at the Edge

Traditional ML inference is evolving toward modeless inference, which abstracts the complexity of model selection from users, allowing the system to automatically choose the most appropriate model for each request based on accuracy and resource requirements. While prior studies have focused on modeless inference within data centers, this paper tackles the pressing need for cost-efficient modeless inference at the edge -- particularly within its unique constraints of limited device memory, volatile network conditions, and restricted power consumption. To overcome these challenges, we propose EdgeSight, a system that provides cost-efficient EdgeSight serving for diverse DNNs at the edge. EdgeSight employs an edge-data center (edge-DC) architecture, utilizing confidence scaling to reduce the number of model options while meeting diverse accuracy requirements. Additionally, it supports lossy inference in volatile network environments. Our experimental results show that EdgeSight outperforms existing systems by up to 1.6x in P99 latency for modeless services. Furthermore, our FPGA prototype demonstrates similar performance at certain accuracy levels, with a power consumption reduction of up to 3.34x.

eess.SY↗

arXiv2024

NetFlowGen: Leveraging Generative Pre-training for Network Traffic Dynamics

Understanding the traffic dynamics in networks is a core capability for automated systems to monitor and analyze networking behaviors, reducing expensive human efforts and economic risks through tasks such as traffic classification, congestion prediction, and attack detection. However, it is still challenging to accurately model network traffic with machine learning approaches in an efficient and broadly applicable manner. Task-specific models trained from scratch are used for different networking applications, which limits the efficiency of model development and generalization of model deployment. Furthermore, while networking data is abundant, high-quality task-specific labels are often insufficient for training individual models. Large-scale self-supervised learning on unlabeled data provides a natural pathway for tackling these challenges. We propose to pre-train a general-purpose machine learning model to capture traffic dynamics with only traffic data from NetFlow records, with the goal of fine-tuning for different downstream tasks with small amount of labels. Our presented NetFlowGen framework goes beyond a proof-of-concept for network traffic pre-training and addresses specific challenges such as unifying network feature representations, learning from large unlabeled traffic data volume, and testing on real downstream tasks in DDoS attack detection. Experiments demonstrate promising results of our pre-training framework on capturing traffic dynamics and adapting to different networking tasks.

cs.LG↗

arXiv2024

NEO: Saving GPU Memory Crisis with CPU Offloading for Online LLM Inference

Online LLM inference powers many exciting applications such as intelligent chatbots and autonomous agents. Modern LLM inference engines widely rely on request batching to improve inference throughput, aiming to make it cost-efficient when running on expensive GPU accelerators. However, the limited GPU memory has largely limited the batch size achieved in practice, leaving significant GPU compute resources wasted. We present NEO, an online LLM inference system that offloads part of attention compute and KV cache states from the GPU to the local host CPU, effectively increasing the GPU batch size and thus inference throughput. To this end, NEO proposes asymmetric GPU-CPU pipelining and load-aware scheduling to balance GPU and CPU loads and fully utilize their compute and memory resources. We evaluate NEO on a wide range of workloads (i.e., code generation, text summarization), GPUs (i.e., T4, A10G, H100), and LLM models (i.e., 7B, 8B, 70B). NEO achieves up to 7.5$\times$, 26%, and 14% higher throughput compared to GPU-only approach on T4, A10G, and H100 GPUs, respectively, while maintaining the same latency; with more powerful CPUs, NEO achieves up to 79.3% throughput gain on A10G GPU.

cs.DC↗

arXiv2024

FRANCIS: Fast Reaction Algorithms for Network Coordination In Switches

Optimizing the reaction to network events, which is critical in tasks such as clock synchronization, multicast, and routing, becomes increasingly challenging as networks grow larger. To improve the reaction time compared to centralized solutions, the theory community has made significant progress in the design of message-passing algorithms that leverage all nodes for distributed computation, and the advent of programmable switches makes it now possible to materialize them. We propose FRANCIS, a framework and associated libraries for running message-passing algorithms on programmable switches. It features primitives that allow easy integration of such algorithms for quickly reacting to network events while optimizing resource consumption. We use FRANCIS to implement event reaction solutions that improve clock synchronization, source-routed multicast, and routing and demonstrate up to 18x reduction in reaction time.

cs.NI↗

arXiv2024

Fast Inference for Augmented Large Language Models

Augmented Large Language Models (LLMs) enhance the capabilities of standalone LLMs by integrating external data sources through API calls. In interactive LLM applications, efficient scheduling is crucial for maintaining low request completion times, directly impacting user engagement. However, these augmentations introduce scheduling challenges due to the need to manage limited memory for cached information (KV caches). As a result, traditional size-based scheduling algorithms, such as Shortest Job First (SJF), become less effective at minimizing completion times. Existing work focuses only on handling requests during API calls by preserving, discarding, or swapping memory without considering how to schedule requests with API calls. In this paper, we propose LAMPS, a novel LLM inference framework for augmented LLMs. LAMPS minimizes request completion time through a unified scheduling approach that considers the total length of requests and their handling strategies during API calls. Recognizing that LLM inference is memory-bound, our approach ranks requests based on their consumption of memory over time, which depends on both the output sizes and how a request is managed during its API calls. To implement our scheduling, LAMPS predicts the strategy that minimizes memory waste of a request during its API calls, aligning with but improving upon existing approaches. We also propose starvation prevention techniques and optimizations to mitigate the overhead of our scheduling. We implement LAMPS on top of vLLM and evaluate its performance against baseline LLM inference systems, demonstrating improvements in end-to-end latency by 27%-85% and reductions in TTFT by 4%-96% compared to the existing augmented-LLM system, with even greater gains over vLLM.

cs.LG↗

arXiv2024

Don't Stop Me Now: Embedding Based Scheduling for LLMs

Efficient scheduling is crucial for interactive Large Language Model (LLM) applications, where low request completion time directly impacts user engagement. Size-based scheduling algorithms like Shortest Remaining Process Time (SRPT) aim to reduce average request completion time by leveraging known or estimated request sizes and allowing preemption by incoming jobs with shorter service times. However, two main challenges arise when applying size-based scheduling to LLM systems. First, accurately predicting output lengths from prompts is challenging and often resource-intensive, making it impractical for many systems. As a result, the state-of-the-art LLM systems default to first-come, first-served scheduling, which can lead to head-of-line blocking and reduced system efficiency. Second, preemption introduces extra memory overhead to LLM systems as they must maintain intermediate states for unfinished (preempted) requests. In this paper, we propose TRAIL, a method to obtain output predictions from the target LLM itself. After generating each output token, we recycle the embedding of its internal structure as input for a lightweight classifier that predicts the remaining length for each running request. Using these predictions, we propose a prediction-based SRPT variant with limited preemption designed to account for memory overhead in LLM systems. This variant allows preemption early in request execution when memory consumption is low but restricts preemption as requests approach completion to optimize resource utilization. On the theoretical side, we derive a closed-form formula for this SRPT variant in an M/G/1 queue model, which demonstrates its potential value. In our system, we implement this preemption policy alongside our embedding-based prediction method.

cs.LG↗

arXiv2024

Carbon Connect: An Ecosystem for Sustainable Computing

Computing is at a moment of profound opportunity. Emerging applications -- such as capable artificial intelligence, immersive virtual realities, and pervasive sensor systems -- drive unprecedented demand for computer. Despite recent advances toward net zero carbon emissions, the computing industry's gross energy usage continues to rise at an alarming rate, outpacing the growth of new energy installations and renewable energy deployments. A shift towards sustainability is needed to spark a transformation in how computer systems are manufactured, allocated, and consumed. Carbon Connect envisions coordinated research thrusts that produce design and management strategies for sustainable, next-generation computer systems. These strategies must flatten and then reverse growth trajectories for computing power and carbon for society's most rapidly growing applications such as artificial intelligence and virtual spaces. We will require accurate models for carbon accounting in computing technology. For embodied carbon, we must re-think conventional design strategies -- over-provisioned monolithic servers, frequent hardware refresh cycles, custom silicon -- and adopt life-cycle design strategies that more effectively reduce, reuse and recycle hardware at scale. For operational carbon, we must not only embrace renewable energy but also design systems to use that energy more efficiently. Finally, new hardware design and management strategies must be cognizant of economic policy and regulatory landscape, aligning private initiatives with societal goals. Many of these broader goals will require computer scientists to develop deep, enduring collaborations with researchers in economics, law, and industrial ecology to spark change in broader practice.

cs.DC↗

arXiv2024

Teal: Learning-Accelerated Optimization of WAN Traffic Engineering

The rapid expansion of global cloud wide-area networks (WANs) has posed a challenge for commercial optimization engines to efficiently solve network traffic engineering (TE) problems at scale. Existing acceleration strategies decompose TE optimization into concurrent subproblems but realize limited parallelism due to an inherent tradeoff between run time and allocation performance. We present Teal, a learning-based TE algorithm that leverages the parallel processing power of GPUs to accelerate TE control. First, Teal designs a flow-centric graph neural network (GNN) to capture WAN connectivity and network flows, learning flow features as inputs to downstream allocation. Second, to reduce the problem scale and make learning tractable, Teal employs a multi-agent reinforcement learning (RL) algorithm to independently allocate each traffic demand while optimizing a central TE objective. Finally, Teal fine-tunes allocations with ADMM (Alternating Direction Method of Multipliers), a highly parallelizable optimization algorithm for reducing constraint violations such as overutilized links. We evaluate Teal using traffic matrices from Microsoft's WAN. On a large WAN topology with >1,700 nodes, Teal generates near-optimal flow allocations while running several orders of magnitude faster than the production optimization engine. Compared with other TE acceleration schemes, Teal satisfies 6--32% more traffic demand and yields 197--625x speedups.

cs.NI↗