arXiv subjects

Shengda Zhu

Publications and source records attributed to Shengda Zhu.

3 recordsLinked to original sources

arXiv2026

Spandana: Reconciling Strict SLOs with Low Cost under Fine-Grained Load Fluctuations

Cloud-based online services face significant sub-second load fluctuations while needing to meet strict Service Level Objectives (SLOs). Cluster operators often over-provision resources to protect SLOs, sacrificing utilization and cost efficiency. Existing reactive and proactive autoscalers, serverless (FaaS) deployments, and VM/FaaS hybrid systems fail to reconcile strict SLO compliance with low cost and high utilization under fine-grained load fluctuation. We introduce Spandana, an architecture that addresses this trade off by decoupling SLO enforcement from cost optimization. A lightweight controller colocated with each application VM enforces SLOs by steering each arriving request between the VM and FaaS. Requests that can meet the SLO stay on the VM; the remaining requests are forwarded to a stock FaaS layer such as AWS Lambda. For cost optimization, Spandana's resource allocator determines the most-efficient VM provisioning by accounting for VM cost, FaaS cost, and traffic volatility, allowing the VM pool to run at high utilization. Our evaluation shows that Spandana maintains strict SLO adherence, achieves 76-86% CPU utilization, and reduces cost by 5-44% over three SOTA baselines.

cs.DC↗

arXiv2026

SWE Context Bench: A Benchmark for Context Learning in Coding

Large language models are increasingly used as coding agents for software engineering tasks. Current benchmarks mainly evaluate whether the agent can correctly solve the request or fix the bugs. They largely treat tasks as independent and do not assess whether agents can reuse previous experience across related problems. As a result, the efficiency gains from reusing the previous experience remains difficult to measure. We introduce SWE-ContextBench, a benchmark designed to explicitly evaluate context understanding and retrieval in coding agents. SWE-ContextBench consists of 1,100 base tasks with another 376 related tasks derived from real dependency and reference relationships among GitHub issues and pull requests. SWE-ContextBench groups base tasks and related tasks with shared context across 51 unique repositories and 9 programming languages. The benchmark evaluates how accurately and efficiently agents solve related issues when prior cases are available in context. Using SWE-ContextBench, we study the behavior of multiple coding agents across varying context reuse settings and retrieval strategies. Our results show that accurately summarized and retrieved previous experience can significantly improve resolution accuracy and reduce runtime and token cost, particularly on harder tasks. In contrast, unfiltered or incorrectly selected context provides limited or negative benefits. These findings highlight the importance of context management and retrieval accuracy, and position SWE-ContextBench as a principled benchmark for studying context learning in coding agents.

cs.SE↗

arXiv2025

Odyssey: An End-to-End System for Pareto-Optimal Serverless Query Processing

Running data analytics queries on serverless (FaaS) workers has been shown to be cost- and performance-efficient for a variety of real-world scenarios, including intermittent query arrival patterns, sudden load spikes and management challenges that afflict managed VM clusters. Alas, existing serverless data analytics works focus primarily on the serverless execution engine and assume the existence of a "good" query execution plan or rely on user guidance to construct such a plan. Meanwhile, even simple analytics queries on serverless have a huge space of possible plans, with vast differences in both performance and cost among plans. This paper introduces Odyssey, an end-to-end serverless-native data analytics pipeline that integrates a query planner, cost model and execution engine. Odyssey automatically generates and evaluates serverless query plans, utilizing state space pruning heuristics and a novel search algorithm to identify Pareto-optimal plans that balance cost and performance with low latency even for complex queries. Our evaluations demonstrate that Odyssey accurately predicts both monetary cost and latency, and consistently outperforms AWS Athena on cost and/or latency.

cs.DB↗