May 2026· Proceedings of the VLDB Endowment· 0 citations· 40 references
Computer ScienceEngineering
TL;DR
This work proposes and develops an open-source policy simulation framework, LoadStar, which forms a reusable benchmark pipeline for validating policies for resource-centric NoSQL workloads, and defines a resource optimization problem for placing Cosmos DB replicas onto VM nodes, and develops the Luna model for forecasting future load distributions.
Abstract
Large-scale managed cloud databases leverage sophisticated load Packing and Migration (PAM) algorithms, which provide the efficiencies necessary for running these services at scale on cloud resources. Research into optimizing the resources and reliability of cloud databases at massive scales is limited by a lack of public NoSQL workloads. We address this in the context of
Cosmos DB
, Microsoft's flagship cloud-hosted NoSQL database. We first propose
open-source NoSQL workloads
from real Cosmos DB clusters, and analyze these traces to derive a novel reliability metric,
Distressed Resource Volume (DRV)
, which captures the quality of service experienced by the end user. We then develop an
open-source policy simulation framework, LoadStar
, powered by a non-parametric statistical model of estimating the QoS of real traffic patterns. These form a reusable benchmark pipeline for validating policies for resource-centric NoSQL workloads. We then define a
resource optimization problem
for placing Cosmos DB replicas onto VM nodes, develop the Luna model for
forecasting
future load distributions, and the Orbit
PAM algorithm
that uses these forecasts to trigger and rebalance stressed replicas, to reduce tail-errors. Our experiments, validated using LoadStar for these workloads, demonstrate Orbit's benefits over the existing Cosmos DB policy and a worst-fit optimized baseline, with higher load delivered at lower error rates and up to 35% reduction in resources. These have been deployed in production, with potential savings of $100
Ms
/yr while improving service reliability for millions of customers.
A comprehensive review of Kubernetes scheduling strategies published between January 2023 and January 2026 is presented and a multi-dimensional taxonomy is established that categorizes scheduling approaches based on common objectives, modification methods, optimization methodologies, targeted workloads, evaluation methods, scheduling scopes, and performance metrics.
Mohammed Alhakimi, R. Latip· Computers· 0 citations
Modern cloud data warehouses decouple compute from storage, making it easy for organizations to access the same underlying data with multiple compute clusters. This flexibility is often used for performance isolation among diverse workloads, so that each workload meets its latency service-level objective (SLO) more reliably. For example, interactive dashboards, ad hoc analysis, and batch jobs can each run on separate clusters. However, this dedicated-cluster approach requires each compute cluster to be continuously scaled to adapt to workload evolution, with over-provisioning wasting resources and under-provisioning risking SLO violations. We present AutoSLO, a latency-SLO-aware workload management framework for multi-cluster cloud data warehouses. AutoSLO operates across three timescales through three key components. First, a periodic Policy Tuner plans proactive cluster scaling actions and tunes configuration parameters, using simulations of history-derived workload forecasts. Second, an SLO-aware reactive Autoscaler adjusts the active cluster set when recent workload behavior deviates from the forecast. Third, an online Query Router reacts to live load when placing each query, using a concurrency-aware latency predictor to avoid SLO violations. On realistic Redbench workloads, AutoSLO successfully meets latency SLOs of varying strictness, reducing cost by a mean of 26.4% compared to the per-scenario next-best baseline. Component-level evaluations show that the Query Router and Autoscaler respectively reduce SLO violation rates by a mean of 47.8% and 93.7%, relative to their corresponding alternatives. Finally, we show that the Policy Tuner can reduce the SLO violation rate by a mean of 44.6% using a single day of workload history, and that each component is efficient given its intended operating timescale.
An Energy-Driven Adaptive Scheduling (EDAS) algorithm is proposed that dynamically prioritizes queries based on estimated CPU utilization, disk I/O costs, and historical energy profiles without requiring modifications to the underlying database engine.
Shankar Kumar· International Journal of Cre...· 0 citations
Function-as-a-Service (FaaS) has revolutionized cloud computing by offering an unprecedented model, with key advantages such as: (1) no infrastructure management, (2) high scalability with fast startup times, and (3) pay-as-you-go pricing model. However, despite its benefits, the high costs of FaaS compared to other computational alternatives lead to increased execution costs for some workloads. This paper presents SpotWhisk, an open-source FaaS-like platform that replaces the expensive computational units of FaaS with spot instances: preemptible virtual machines offered by the provider, which are up to 87.4% more economical than traditional FaaS computational units. SpotWhisk, which maintains the same programming model and infrastructure abstraction as FaaS, allows users to easily migrate their serverless data analytics pipelines and schedule new functions using the same simplified model. Although SpotWhisk introduces the overhead of initializing the underlying spot instances, we demonstrate, through real-world workloads, that: (1) we achieve a very low eviction rate for FaaS workloads, (2) we achieve savings of up to 84% in execution costs, and (3) we improve the overall performance (time-to-cost ratio) of the evaluated applications by up to 6.15×.
Enrique Molina-Giménez, Pedro García-López, Javier Fabra· IEEE International Conferenc...· 0 citations
The new EMC+ proposal is an OS‐driven elasticity manager for container‐based environments that continuously estimates idle core cycles left by regular (inelastic) applications, and reallocates idle cores to elastic ones, even during short time intervals, and has minimal impact on the performance and QoS of colocated inelastic applications.
J. C. Saez, Carlos Bilbao, Manuel Prieto-Matías· Concurrency and Computation· 0 citations
