Cold Start Latency in Serverless Computing: A Systematic Review, Taxonomy, and Future Directions

ACM Computing Surveys, Год журнала: 2024, Номер 57(3), С. 1 - 36

Опубликована: Окт. 17, 2024

Recently, academics and the corporate sector have paid attention to serverless computing, which enables dynamic scalability an economic model. In users only pay for time they actually use resources, enabling zero scaling optimise cost resource utilisation. However, this approach also introduces cold start problem. Researchers developed various solutions address problem, yet it remains unresolved research area. article, we propose a systematic literature review on latency in computing. Furthermore, create detailed taxonomy of approaches latency, investigate existing techniques reducing frequency. We classified current studies into several categories such as caching application-level optimisation-based solutions, well Artificial Intelligence/Machine Learning-based solutions. Moreover, analyzed impact quality service, explored mitigation methods, datasets, implementation platforms, them based their common characteristics features. Finally, outline open challenges highlight possible future directions.

Язык: Английский

---

Scalable compute continuum

Future Generation Computer Systems, Год журнала: 2025, Номер unknown, С. 107697 - 107697

Опубликована: Янв. 1, 2025

---

An extensible lightweight framework for distributed telemetry of microservices

Sustainable Computing Informatics and Systems, Год журнала: 2025, Номер 46, С. 101100 - 101100

Опубликована: Фев. 26, 2025

Язык: Английский

---

HEROES: Humanitarian Emergency Response based on UAV-enabled Integrated Sensing and Communication, Positioning, and Satisfaction Games

ACM Journal on Autonomous Transportation Systems, Год журнала: 2025, Номер unknown

Опубликована: Фев. 24, 2025

In the field of autonomous transportation systems, integration Unmanned Aerial Vehicles (UAVs) in emergency response scenarios is important for enhancing operational efficiency and victims’ positioning. This paper presents a novel Positioning, Navigation, Timing (PNT) framework, named HEROES, which leverages UAV Integrated Sensing Communication (ISAC) technologies to address challenges post-disaster environments. Our approach focuses on comprehensive scenario involving multiple victims, first responders, UAVs, an Emergency Control Center (ECC). HEROES enables UAVs function as anchor nodes facilitate precise positioning victims while simultaneously collecting critical data from disaster area. We further introduce Reinforcement Learning (RL) model based Optimistic Q-learning with Upper Bound Confidence algorithm, enabling responders autonomously select most advantageous connections their channel gain, shadowing probability, positional characteristics. Furthermore, Satisfaction game-theoretic enhance sensing, communication, functionalities. analysis reveals existence various satisfaction equilibria, including Minimum Efficient Equilibrium (MESE), ensuring that meet Quality Service (QoS) constraints at minimal costs. Extensive experimental results validate scalability performance demonstrating significant improvements over existing state-of-the-art methods delivering PNT services during humanitarian emergencies.

Язык: Английский

---

Active Inference on the Edge: A Design Study

2022 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops), Год журнала: 2024, Номер unknown, С. 550 - 555

Опубликована: Март 11, 2024

Every year, the amount of data created by Internet Things (IoT) devices increases; therefore, processing is carried out edge in close proximity. To ensure Quality Service (QoS) throughout these operations, systems are supervised and adapted with help Machine Learning (ML). However, as long ML models not retrained, they fail to capture gradual shifts variable distribution, leading an inaccurate view system state poor inference. In this paper, we present a novel paradigm that constructed upon Active Inference (ACI) – concept from neuroscience describes how brain constantly predicts evaluates sensory information decrease long-term surprise. We implemented use case, which ACI-based agent continuously optimized operation on smart manufacturing engine according QoS requirements. The used causal knowledge gradually develop understanding its actions related requirements fulfillment, configurations favor. As result, our required 5 cycles converge optimal solution.

Язык: Английский

---

Markov Blanket Composition of SLOs

Опубликована: Июль 7, 2024

Язык: Английский

---

Diffusing High-level SLO in Microservice Pipelines

Опубликована: Июль 15, 2024

Язык: Английский

---

SLO-Aware Task Offloading Within Collaborative Vehicle Platoons

Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 72 - 86

Опубликована: Дек. 6, 2024

Язык: Английский

---