arXiv subjects

Qingda Lu

Publications and source records attributed to Qingda Lu.

5 recordsLinked to original sources

arXiv2026

Pythia: Exploiting Workflow Predictability for Efficient Agent-Native LLM Serving

As LLM applications grow more complex, developers are increasingly adopting multi-agent architectures to decompose workflows into specialized, collaborative components, introducing structure that constrains agent behavior and exposes useful semantic predictability. Unlike traditional LLM serving, which operates under highly dynamic and uncertain conditions, this structured topology enables opportunities to reduce runtime uncertainty$\unicode{x2015}$yet existing systems fail to exploit it, treating agentic workloads as generic traffic and incurring significant inefficiencies. Our analysis of production traces from an agent-serving platform and an internal coding assistant reveals key bottlenecks, including low prefix cache hit rates, severe resource contention from long-context requests, and substantial queuing delays due to suboptimal scaling. To address these challenges, we propose Pythia, a multi-agent serving system that captures workflow semantics through a simple interface at the serving layer, unlocking new optimization opportunities and substantially improving throughput and job completion time over state-of-the-art baselines.

cs.MA↗

arXiv2025

Prism: Cost-Efficient Multi-LLM Serving via GPU Memory Ballooning

Inference providers must maintain availability for many LLMs, including low-volume but essential models, making resource efficiency increasingly important as token prices fall. Analysis of production traces reveals a dynamic bursty-group pattern in which sets of models become active together and shift over time; existing space- and time-sharing approaches lack principled mechanisms to adapt to this variability, forcing trade-offs between SLO adherence and efficiency. We observe that elastic memory allocation can unify spatial and temporal sharing. Based on this insight, we have developed Prism, a memory-centric LLM co-serving framework that applies memory ballooning to reclaim memory across models and support both forms of sharing under a single scheme. Prism's balloon driver, referred to as kvcached, has been open-sourced at https://github.com/ovg-project/kvcached, and deployed in production environments across 10K+ GPUs.

cs.DC↗

arXiv2024

Scaling Inter-procedural Dataflow Analysis on the Cloud

Apart from forming the backbone of compiler optimization, static dataflow analysis has been widely applied in a vast variety of applications, such as bug detection, privacy analysis, program comprehension, etc. Despite its importance, performing interprocedural dataflow analysis on large-scale programs is well known to be challenging. In this paper, we propose a novel distributed analysis framework supporting the general interprocedural dataflow analysis. Inspired by large-scale graph processing, we devise dedicated distributed worklist algorithms for both whole-program analysis and incremental analysis. We implement these algorithms and develop a distributed framework called BigDataflow running on a large-scale cluster. The experimental results validate the promising performance of BigDataflow -- BigDataflow can finish analyzing the program of millions lines of code in minutes. Compared with the state-of-the-art, BigDataflow achieves much more analysis efficiency.

cs.PL↗

arXiv2024

LogParser-LLM: Advancing Efficient Log Parsing with Large Language Models

Logs are ubiquitous digital footprints, playing an indispensable role in system diagnostics, security analysis, and performance optimization. The extraction of actionable insights from logs is critically dependent on the log parsing process, which converts raw logs into structured formats for downstream analysis. Yet, the complexities of contemporary systems and the dynamic nature of logs pose significant challenges to existing automatic parsing techniques. The emergence of Large Language Models (LLM) offers new horizons. With their expansive knowledge and contextual prowess, LLMs have been transformative across diverse applications. Building on this, we introduce LogParser-LLM, a novel log parser integrated with LLM capabilities. This union seamlessly blends semantic insights with statistical nuances, obviating the need for hyper-parameter tuning and labeled training data, while ensuring rapid adaptability through online parsing. Further deepening our exploration, we address the intricate challenge of parsing granularity, proposing a new metric and integrating human interactions to allow users to calibrate granularity to their specific needs. Our method's efficacy is empirically demonstrated through evaluations on the Loghub-2k and the large-scale LogPub benchmark. In evaluations on the LogPub benchmark, involving an average of 3.6 million logs per dataset across 14 datasets, our LogParser-LLM requires only 272.5 LLM invocations on average, achieving a 90.6% F1 score for grouping accuracy and an 81.1% for parsing accuracy. These results demonstrate the method's high efficiency and accuracy, outperforming current state-of-the-art log parsers, including pattern-based, neural network-based, and existing LLM-enhanced approaches.

cs.SE↗

arXiv2024

DRust: Language-Guided Distributed Shared Memory with Fine Granularity, Full Transparency, and Ultra Efficiency

Despite being a powerful concept, distributed shared memory (DSM) has not been made practical due to the extensive synchronization needed between servers to implement memory coherence. This paper shows a practical DSM implementation based on the insight that the ownership model embedded in programming languages such as Rust automatically constrains the order of read and write, providing opportunities for significantly simplifying the coherence implementation if the ownership semantics can be exposed to and leveraged by the runtime. This paper discusses the design and implementation of DistR, a Rust-based DSM system that outperforms the two state-of-the-art DSM systems GAM and Grappa by up to 2.64x and 29.16x in throughput, and scales much better with the number of servers.

cs.DC↗