15 papers
Online Batching with Lookahead: Adversarial Limits and Stochastic Gains
Dimitris Christou, Jiayi Chen, Nihal Sharma +3
Modern inference systems rely on batching to exploit parallel execution and amortize serial execution overheads, but batching creates a fundamental online scheduling tradeoff: serv…
AtumAI: A Principled Framework for Agentic Generation of Datacenter Control-Plane Policies
Qiushi Lin, Chaojie Zhang, Íñigo Goiri +3
The efficiency of a datacenter rests on its control plane policies. Designing these policies is increasingly hard: the hardware-software stack grows fast, the design space is vast…
AutoScout: Structured Optimization for Automating ML System Configuration
Jimmy Shong, Yuhan Ding, Yihan Jiang +5
Machine learning (ML) systems expose a rapidly expanding configuration space spanning model-parallelism strategies, communication optimizations, and low-level runtime parameters. E…
Vulcan: Instance-specialized, Verifiable Systems Heuristics Through LLM-driven Search
Rohit Dwivedula, Divyanshu Saxena, Sujay Yadalam +3
Systems resource management tasks rely primarily on hand-designed heuristics. However, growing hardware heterogeneity and workload diversity require heuristics specialized to parti…
Continuous Benchmark Generation for Evaluating Enterprise-scale LLM Agents
Divyanshu Saxena, Rishikesh Maurya, Xiaoxuan Ou +7
The rapid adoption of AI agents across domains has made systematic evaluation crucial for ensuring their usefulness and successful production deployment. Evaluation of AI agents ty…
A Joint Learning Approach to Hardware Caching and Prefetching
Samuel Yuan, Divyanshu Saxena, Jiayi Chen +2
Several learned policies have been proposed to replace heuristics for scheduling, caching, and other system components in modern systems. By leveraging diverse features, learning f…
Man-Made Heuristics Are Dead. Long Live Code Generators!
Rohit Dwivedula, Divyanshu Saxena, Aditya Akella +2
Policy design for various systems controllers has conventionally been a manual process, with domain experts carefully tailoring heuristics for the specific instance in which the po…
ConfigBot: Adaptive Resource Allocation for Robot Applications in Dynamic Environments
Rohit Dwivedula, Sadanand Modak, Aditya Akella +3
The growing use of service robots in dynamic environments requires flexible management of on-board compute resources to optimize the performance of diverse tasks such as navigation…
Large Language Models as Realistic Microservice Trace Generators
Donghyun Kim, Sriram Ravula, Taemin Ha +3
Workload traces are essential to understand complex computer systems' behavior and manage processing and memory resources. Since real-world traces are hard to obtain, synthetic tra…
Reforge: Low-Latency Distributed GNN Serving with Selective Embedding Recomputation
Geon-Woo Kim, Donghyun Kim, Jeongyoon Moon +5
Graph Neural Networks (GNNs) have been widely adopted for their ability to compute expressive node representations in graph datasets. However, serving GNNs on large graphs is chall…
Canopy: Property-Driven Learning for Congestion Control
Chenxi Yang, Divyanshu Saxena, Rohit Dwivedula +3
Learning-based congestion controllers offer better adaptability compared to traditional heuristics. However, the unreliability of learning techniques can cause learning-based contr…
CONGO: Compressive Online Gradient Optimization
Jeremy Carleton, Prathik Vijaykumar, Divyanshu Saxena +3
We address the challenge of zeroth-order online convex optimization where the objective function's gradient exhibits sparsity, indicating that only a small number of dimensions pos…
Accelerating Distributed Deep Learning using Lossless Homomorphic Compression
Haoyu Li, Yuchen Xu, Jiayi Chen +5
As deep neural networks (DNNs) grow in complexity and size, the resultant increase in communication overhead during distributed training has become a significant bottleneck, challe…
On a Foundation Model for Operating Systems
Divyanshu Saxena, Nihal Sharma, Donghyun Kim +15
This paper lays down the research agenda for a domain-specific foundation model for operating systems (OSes). Our case for a foundation model revolves around the observations that…
Multi-agent Databases via Independent Learning
Chi Zhang, Olga Papaemmanouil, Josiah P. Hanna +1
Machine learning is rapidly being used in database research to improve the effectiveness of numerous tasks included but not limited to query optimization, workload scheduling, phys…