Nalar: An agent serving framework

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Laju, Marco, Son, Donghyun, Agarwal, Saurabh, Kedia, Nitin, Lee, Myungjin, Srinivasa, Jayanth, Akella, Aditya
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915716975820800
author Laju, Marco
Son, Donghyun
Agarwal, Saurabh
Kedia, Nitin
Lee, Myungjin
Srinivasa, Jayanth
Akella, Aditya
author_facet Laju, Marco
Son, Donghyun
Agarwal, Saurabh
Kedia, Nitin
Lee, Myungjin
Srinivasa, Jayanth
Akella, Aditya
contents LLM-driven agentic applications increasingly automate complex, multi-step tasks, but serving them efficiently remains challenging due to heterogeneous components, dynamic and model-driven control flow, long-running state, and unpredictable latencies. Nalar is a ground-up agent-serving framework that cleanly separates workflow specification from execution while providing the runtime visibility and control needed for robust performance. Nalar preserves full Python expressiveness, using lightweight auto-generated stubs that turn agent and tool invocations into futures carrying dependency and context metadata. A managed state layer decouples logical state from physical placement, enabling safe reuse, migration, and consistent retry behavior. A two-level control architecture combines global policy computation with local event-driven enforcement to support adaptive routing, scheduling, and resource management across evolving workflows. Together, these mechanisms allow Nalar to deliver scalable, efficient, and policy-driven serving of heterogeneous agentic applications without burdening developers with orchestration logic. Across three agentic workloads, Nalar cuts tail latency by 34--74\%, achieves up to $2.9\times$ speedups, sustains 80 RPS where baselines fail, and scales to 130K futures with sub-500 ms control overhead.
format Preprint
id arxiv_https___arxiv_org_abs_2601_05109
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Nalar: An agent serving framework
Laju, Marco
Son, Donghyun
Agarwal, Saurabh
Kedia, Nitin
Lee, Myungjin
Srinivasa, Jayanth
Akella, Aditya
Distributed, Parallel, and Cluster Computing
Multiagent Systems
LLM-driven agentic applications increasingly automate complex, multi-step tasks, but serving them efficiently remains challenging due to heterogeneous components, dynamic and model-driven control flow, long-running state, and unpredictable latencies. Nalar is a ground-up agent-serving framework that cleanly separates workflow specification from execution while providing the runtime visibility and control needed for robust performance. Nalar preserves full Python expressiveness, using lightweight auto-generated stubs that turn agent and tool invocations into futures carrying dependency and context metadata. A managed state layer decouples logical state from physical placement, enabling safe reuse, migration, and consistent retry behavior. A two-level control architecture combines global policy computation with local event-driven enforcement to support adaptive routing, scheduling, and resource management across evolving workflows. Together, these mechanisms allow Nalar to deliver scalable, efficient, and policy-driven serving of heterogeneous agentic applications without burdening developers with orchestration logic. Across three agentic workloads, Nalar cuts tail latency by 34--74\%, achieves up to $2.9\times$ speedups, sustains 80 RPS where baselines fail, and scales to 130K futures with sub-500 ms control overhead.
title Nalar: An agent serving framework
topic Distributed, Parallel, and Cluster Computing
Multiagent Systems
url https://arxiv.org/abs/2601.05109