Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads
For system designers building scalable LLM agents, this work provides foundational understanding and practical guidance for optimizing memory systems in long-horizon tasks.
This paper presents the first systems characterization of LLM agent memory, introducing a taxonomy and profiling harness to analyze ten representative systems across two benchmarks, revealing how design choices shift costs between write and read paths, and deriving ten system recommendations.
LLM agents are increasingly deployed on long-horizon tasks requiring sustained reasoning over extended interaction histories. Realizing this at scale requires agents to persistently store, retrieve, and update their own memory across sessions. A rich ecosystem of agent memory systems has emerged spanning flat retrieval, LLM-mediated extraction, consolidating fact stores, and agentic control flows. Yet, their system-level behavior remains uncharacterized. We present the first systems characterization of agent memory. First, we introduce a system-oriented taxonomy classifying agent memory systems along four axes. Second, we build a phase-aware profiling harness attributing cost to construction, retrieval, and generation. Third, we characterize ten representative systems across two benchmark suites, uncovering how design choices shift cost across the write and read paths. Finally, we derive 10 system recommendations covering construction scheduling, capability floors, amortization via query volume, freshness-latency tradeoffs, and fleet-scale management.