Skip to content

The Dataflow Model Revisited

· 0 citations · 65 references

TL;DR

It is explored how the completeness principle split into two successful forms: watermarks and snapshot-consistent refresh (where streams stay visible) and snapshot-consistent refresh (where they do not), and why the latter reached far more users by asking far less of them, and generalize the former into declared constraints on change.

View source

Similar papers

Preprint Sep 2026

ChurnBench: A Drift-Aware Benchmark Demonstrating That Refresh Scheduling, Not Cache Age, Governs Staleness in Agentic AI

In production, agentic systems answer questions over data that lives in several places and keeps changing: licenses are reassigned, users offboarded, prices changed, contracts renewed. Existing retrieval benchmarks freeze the data, so they cannot ask whether an agent's answer is still true, only whether it found the ri...

V. Singh, Preeti Priyam, Gautam Bhowmick · 0 citations
Jul 2026

Fugue: Online Elasticity for Distributed Stateful Stream Processing

Fugue is a novel, self-contained reactive protocol that provides seamless and resource-efficient elasticity and reaches comparable handover performance while avoiding continuous replication overhead, and is implemented in Apache Flink.

Yu-Qiu Zhang, Yun-Hao Mao, Hans-Arno Jacobsen · 0 citations
Preprint Aug 2026

LLM within MCP Matters: Measuring Inefficient Resource Utilization Driven by LLMs

Per-server prompt engineering is therefore a workaround rather than a fix; it is argued that MCP host applications should provide an explicit mechanism that places server instructions ahead of tool selection in the client LLM's deliberation.

Minhan Cho, Soyoung Park, Kihyeon Jeong et al. · 0 citations
Preprint Aug 2026

Composable Building Blocks for Resilient Asynchronous Code

This work shows how higher-order combinators solve higher-order problems of asynchronous calls to a network service, database, or language model uniformly, including timeouts, retries, rate limiting, caching, reentrant locking, and cancellation.

Frank Tip · 0 citations
Book Open access Sep 2026

LLM KV-cache: To Restore or To Recompute, That Is the Question

The challenges of the restoration-recomputation trade-off are investigated and its impact on inference performance when left unaddressed, and an I/O-aware KV-cache management policy is presented that dynamically navigates this trade-off.

Amirhossein Najafizadeh, Vasily Tarasov, Alex Merenstein et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.