arXiv subjects

Kejiang Ye

Publications and source records attributed to Kejiang Ye.

At least 55 records · Page 3Linked to original sources

arXiv2024

LSRAM: A Lightweight Autoscaling and SLO Resource Allocation Framework for Microservices Based on Gradient Descent

Microservices architecture has become the dominant architecture in cloud computing paradigm with its advantages of facilitating development, deployment, modularity and scalability. The workflow of microservices architecture is transparent to the users, who are concerned with the quality of service (QoS). Taking Service Level Objective (SLO) as an important indicator of system resource scaling can effectively ensure user's QoS, but how to quickly allocate end-to-end SLOs to each microservice in a complete service so that it can obtain the optimal SLO resource allocation scheme is still a challenging problem. Existing microservice autoscaling frameworks based on SLO resources often have heavy and complex models that demand substantial time and computational resources to get a suitable resource allocation scheme. Moreover, when the system environment or microservice application changes, these methods require significant time and resources for model retraining. In this paper, we propose LSRAM, a lightweight SLO resource allocation management framework based on the gradient descent method to overcome the limitation of existing methods in terms of heavy model, time-consuming, poor scalability, and difficulty in retraining. LSRAM has two stages: at stage one, the lightweight SLO resource allocation model from LSRAM can quickly compute the appropriate SLO resources for each microservice; at stage two, LSRAM's SLO resource update model enables the entire framework to quickly adapt to changes in the cluster environment (e.g. load and applications). Additionally, LSRAM can effectively handle bursty traffic and highly fluctuating load application scenarios. Compared to state-of-the-art SLO allocation frameworks, LSRAM not only guarantees users' QoS but also reduces resource usage by 17%.

cs.DC↗

arXiv2024

Edge AI: A Taxonomy, Systematic Review and Future Directions

Edge Artificial Intelligence (AI) incorporates a network of interconnected systems and devices that receive, cache, process, and analyze data in close communication with the location where the data is captured with AI technology. Recent advancements in AI efficiency, the widespread use of Internet of Things (IoT) devices, and the emergence of edge computing have unlocked the enormous scope of Edge AI. Edge AI aims to optimize data processing efficiency and velocity while ensuring data confidentiality and integrity. Despite being a relatively new field of research from 2014 to the present, it has shown significant and rapid development over the last five years. This article presents a systematic literature review for Edge AI to discuss the existing research, recent advancements, and future research directions. We created a collaborative edge AI learning system for cloud and edge computing analysis, including an in-depth study of the architectures that facilitate this mechanism. The taxonomy for Edge AI facilitates the classification and configuration of Edge AI systems while examining its potential influence across many fields through compassing infrastructure, cloud computing, fog computing, services, use cases, ML and deep learning, and resource management. This study highlights the significance of Edge AI in processing real-time data at the edge of the network. Additionally, it emphasizes the research challenges encountered by Edge AI systems, including constraints on resources, vulnerabilities to security threats, and problems with scalability. Finally, this study highlights the potential future research directions that aim to address the current limitations of Edge AI by providing innovative solutions.

cs.DC↗

arXiv2024

UELLM: A Unified and Efficient Approach for LLM Inference Serving

In the context of Machine Learning as a Service (MLaaS) clouds, the extensive use of Large Language Models (LLMs) often requires efficient management of significant query loads. When providing real-time inference services, several challenges arise. Firstly, increasing the number of GPUs may lead to a decrease in inference speed due to heightened communication overhead, while an inadequate number of GPUs can lead to out-of-memory errors. Secondly, different deployment strategies need to be evaluated to guarantee optimal utilization and minimal inference latency. Lastly, inefficient orchestration of inference queries can easily lead to significant Service Level Objective (SLO) violations. Lastly, inefficient orchestration of inference queries can easily lead to significant Service Level Objective (SLO) violations. To address these challenges, we propose a Unified and Efficient approach for Large Language Model inference serving (UELLM), which consists of three main components: 1) resource profiler, 2) batch scheduler, and 3) LLM deployer. UELLM minimizes resource overhead, reduces inference latency, and lowers SLO violation rates. Compared with state-of-the-art (SOTA) techniques, UELLM reduces the inference latency by 72.3% to 90.3%, enhances GPU utilization by 1.2X to 4.1X, and increases throughput by 1.92X to 4.98X, it can also serve without violating the inference latency SLO.

cs.DC↗

arXiv2024

MSARS: A Meta-Learning and Reinforcement Learning Framework for SLO Resource Allocation and Adaptive Scaling for Microservices

Service Level Objectives (SLOs) aim to set threshold for service time in cloud services to ensure acceptable quality of service (QoS) and user satisfaction. Currently, many studies consider SLOs as a system resource to be allocated, ensuring QoS meets the SLOs. Existing microservice auto-scaling frameworks that rely on SLO resources often utilize complex and computationally intensive models, requiring significant time and resources to determine appropriate resource allocation. This paper aims to rapidly allocate SLO resources and minimize resource costs while ensuring application QoS meets the SLO requirements in a dynamically changing microservice environment. We propose MSARS, a framework that leverages meta-learning to quickly derive SLO resource allocation strategies and employs reinforcement learning for adaptive scaling of microservice resources. It features three innovative components: First, MSARS uses graph convolutional networks to predict the most suitable SLO resource allocation scheme for the current environment. Second, MSARS utilizes meta-learning to enable the graph neural network to quickly adapt to environmental changes ensuring adaptability in highly dynamic microservice environments. Third, MSARS generates auto-scaling policies for each microservice based on an improved Twin Delayed Deep Deterministic Policy Gradient (TD3) model. The adaptive auto-scaling policy integrates the SLO resource allocation strategy into the scheduling algorithm to satisfy SLOs. Finally, we compare MSARS with state-of-the-art resource auto-scaling algorithms that utilize neural networks and reinforcement learning, MSARS takes 40% less time to adapt to new environments, 38% reduction of SLO violations, and 8% less resources cost.

cs.DC↗

arXiv2024

DRPC: Distributed Reinforcement Learning Approach for Scalable Resource Provisioning in Container-based Clusters

Microservices have transformed monolithic applications into lightweight, self-contained, and isolated application components, establishing themselves as a dominant paradigm for application development and deployment in public clouds such as Google and Alibaba. Autoscaling emerges as an efficient strategy for managing resources allocated to microservices' replicas. However, the dynamic and intricate dependencies within microservice chains present challenges to the effective management of scaled microservices. Additionally, the centralized autoscaling approach can encounter scalability issues, especially in the management of large-scale microservice-based clusters. To address these challenges and enhance scalability, we propose an innovative distributed resource provisioning approach for microservices based on the Twin Delayed Deep Deterministic Policy Gradient algorithm. This approach enables effective autoscaling decisions and decentralizes responsibilities from a central node to distributed nodes. Comparative results with state-of-the-art approaches, obtained from a realistic testbed and traces, indicate that our approach reduces the average response time by 15% and the number of failed requests by 24%, validating improved scalability as the number of requests increases.

cs.DC↗

arXiv2024

StatuScale: Status-aware and Elastic Scaling Strategy for Microservice Applications

Microservice architecture has transformed traditional monolithic applications into lightweight components. Scaling these lightweight microservices is more efficient than scaling servers. However, scaling microservices still faces the challenges resulted from the unexpected spikes or bursts of requests, which are difficult to detect and can degrade performance instantaneously. To address this challenge and ensure the performance of microservice-based applications, we propose a status-aware and elastic scaling framework called StatuScale, which is based on load status detector that can select appropriate elastic scaling strategies for differentiated resource scheduling in vertical scaling. Additionally, StatuScale employs a horizontal scaling controller that utilizes comprehensive evaluation and resource reduction to manage the number of replicas for each microservice. We also present a novel metric named correlation factor to evaluate the resource usage efficiency. Finally, we use Kubernetes, an open-source container orchestration and management platform, and realistic traces from Alibaba to validate our approach. The experimental results have demonstrated that the proposed framework can reduce the average response time in the Sock-Shop application by 8.59% to 12.34%, and in the Hotel-Reservation application by 7.30% to 11.97%, decrease service level objective violations, and offer better performance in resource usage compared to baselines.

cs.DC↗

arXiv2024

Collaborative Resource Management and Workloads Scheduling in Cloud-Assisted Mobile Edge Computing across Timescales

Due to the limited resource capacity of edge servers and the high purchase costs of edge resources, service providers are facing the new challenge of how to take full advantage of the constrained edge resources for Internet of Things (IoT) service hosting and task scheduling to maximize system performance. In this paper, we study the joint optimization problem on service placement, resource provisioning, and workloads scheduling under resource and budget constraints, which is formulated as a mixed integer non-linear programming problem. Given that the frequent service placement and resource provisioning will significantly increase system configuration costs and instability, we propose a two-timescale framework for resource management and workloads scheduling, named RMWS. RMWS consists of a Gibbs sampling algorithm and an alternating minimization algorithm to determine the service placement and resource provisioning on large timescales. And a sub-gradient descent method has been designed to solve the workload scheduling challenge on small timescales.We conduct comprehensive experiments under different parameter settings. The RMWS consistently ensures a minimum 10% performance enhancement compared to other algorithms, showcasing its superiority. Theoretical proofs are also provided accordingly.

cs.DC↗

arXiv2024

TempoScale: A Cloud Workloads Prediction Approach Integrating Short-Term and Long-Term Information

Cloud native solutions are widely applied in various fields, placing higher demands on the efficient management and utilization of resource platforms. To achieve the efficiency, load forecasting and elastic scaling have become crucial technologies for dynamically adjusting cloud resources to meet user demands and minimizing resource waste. However, existing prediction-based methods lack comprehensive analysis and integration of load characteristics across different time scales. For instance, long-term trend analysis helps reveal long-term changes in load and resource demand, thereby supporting proactive resource allocation over longer periods, while short-term volatility analysis can examine short-term fluctuations in load and resource demand, providing support for real-time scheduling and rapid response. In response to this, our research introduces TempoScale, which aims to enhance the comprehensive understanding of temporal variations in cloud workloads, enabling more intelligent and adaptive decision-making for elastic scaling. TempoScale utilizes the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise algorithm to decompose time-series load data into multiple Intrinsic Mode Functions (IMF) and a Residual Component (RC). First, we integrate the IMF, which represents both long-term trends and short-term fluctuations, into the time series prediction model to obtain intermediate results. Then, these intermediate results, along with the RC, are transferred into a fully connected layer to obtain the final result. Finally, this result is fed into the resource management system based on Kubernetes for resource scaling. Our proposed approach can reduce the Mean Square Error by 5.80% to 30.43% compared to the baselines, and reduce the average response time by 5.58% to 31.15%.

cs.DC↗

arXiv2024

An Interference-aware Approach for Co-located Container Orchestration with Novel Metric

Container orchestration technologies are widely employed in cloud computing, facilitating the co-location of online and offline services on the same infrastructure. Online services demand rapid responsiveness and high availability, whereas offline services require extensive computational resources. However, this mixed deployment can lead to resource contention, adversely affecting the performance of online services, yet the metrics used by existing methods cannot accurately reflect the extent of interference. In this paper, we introduce scheduling latency as a novel metric for quantifying interference and compare it with existing metrics. Empirical evidence demonstrates that scheduling latency more accurately reflects the performance degradation of online services. We also utilize various machine learning techniques to predict potential interference on specific hosts for online services, providing reference information for subsequent scheduling decisions. Simultaneously, we propose a method for quantifying node interference based on scheduling latency. To enhance resource utilization, we train a model for online services that predicts CPU and MEM (memory) resource allocation based on workload type and QPS. Finally, we present a scheduling algorithm based on predictive modeling, aiming to reduce interference in online services while balancing node resource utilization. Through experiments and comparisons with three other baseline methods, we demonstrate the effectiveness of our approach. Compared with three baselines, our approach can reduce the average response time, 90th percentile response time, and 99th percentile response time of online services by 29.4%, 31.4%, and 14.5%, respectively.

cs.DC↗

arXiv2023

ChainsFormer: A Chain Latency-aware Resource Provisioning Approach for Microservices Cluster

The trend towards transitioning from monolithic applications to microservices has been widely embraced in modern distributed systems and applications. This shift has resulted in the creation of lightweight, fine-grained, and self-contained microservices. Multiple microservices can be linked together via calls and inter-dependencies to form complex functions. One of the challenges in managing microservices is provisioning the optimal amount of resources for microservices in the chain to ensure application performance while improving resource usage efficiency. This paper presents ChainsFormer, a framework that analyzes microservice inter-dependencies to identify critical chains and nodes, and provision resources based on reinforcement learning. To analyze chains, ChainsFormer utilizes light-weight machine learning techniques to address the dynamic nature of microservice chains and workloads. For resource provisioning, a reinforcement learning approach is used that combines vertical and horizontal scaling to determine the amount of allocated resources and the number of replicates. We evaluate the effectiveness of ChainsFormer using realistic applications and traces on a real testbed based on Kubernetes. Our experimental results demonstrate that ChainsFormer can reduce response time by up to 26% and improve processed requests per second by 8% compared with state-of-the-art techniques.

cs.DC↗

arXiv2023

Practice of Alibaba Cloud on Elastic Resource Provisioning for Large-scale Microservices Cluster

Cloud-native architecture is becoming increasingly crucial for today's cloud computing environments due to the need for speed and flexibility in developing applications. It utilizes microservice technology to break down traditional monolithic applications into light-weight and self-contained microservice components. However, as microservices grow in scale and have dynamic inter-dependencies, they also pose new challenges in resource provisioning that cannot be fully addressed by traditional resource scheduling approaches. The various microservices with different resource needs and latency requirements can create complex calling chains, making it difficult to provide fine-grained and accurate resource allocation to each component while maintaining the overall quality of service in the chain. In this work, we aim to address the research problem on how to efficiently provision resources for the growing scale of microservice platform and ensure the performance of latency-critical microservices. To address the problem, we present in-depth analyses of Alibaba's microservice cluster and propose optimized resource provisioning algorithms to enhance resource utilization while ensuring the latency requirement. First, we analyze the distinct features of microservices in Alibaba's cluster compared to traditional applications. Then we present Alibaba's resource capacity provisioning workflow and framework to address challenges in resource provisioning for large-scale and latency-critical microservice clusters. Finally, we propose enhanced resource provisioning algorithms over Alibaba's current practice by making both proactive and reactive scheduling decisions based on different workloads patterns, which can improve resource usage by 10-15% in Alibaba's clusters, while maintaining the necessary latency for microservices.

cs.DC↗

arXiv2023

APBench: A Unified Benchmark for Availability Poisoning Attacks and Defenses

The efficacy of availability poisoning, a method of poisoning data by injecting imperceptible perturbations to prevent its use in model training, has been a hot subject of investigation. Previous research suggested that it was difficult to effectively counteract such poisoning attacks. However, the introduction of various defense methods has challenged this notion. Due to the rapid progress in this field, the performance of different novel methods cannot be accurately validated due to variations in experimental setups. To further evaluate the attack and defense capabilities of these poisoning methods, we have developed a benchmark -- APBench for assessing the efficacy of adversarial poisoning. APBench consists of 9 state-of-the-art availability poisoning attacks, 8 defense algorithms, and 4 conventional data augmentation techniques. We also have set up experiments with varying different poisoning ratios, and evaluated the attacks on multiple datasets and their transferability across model architectures. We further conducted a comprehensive evaluation of 2 additional attacks specifically targeting unsupervised models. Our results reveal the glaring inadequacy of existing attacks in safeguarding individual privacy. APBench is open source and available to the deep learning community: https://github.com/lafeat/apbench.

cs.CV↗

arXiv2023

Resource Management for GPT-based Model Deployed on Clouds: Challenges, Solutions, and Future Directions

The widespread adoption of the large language model (LLM), e.g. Generative Pre-trained Transformer (GPT), deployed on cloud computing environment (e.g. Azure) has led to a huge increased demand for resources. This surge in demand poses significant challenges to resource management in clouds. This paper aims to highlight these challenges by first identifying the unique characteristics of resource management for the GPT-based model. Building upon this understanding, we analyze the specific challenges faced by resource management in the context of GPT-based model deployed on clouds, and propose corresponding potential solutions. To facilitate effective resource management, we introduce a comprehensive resource management framework and present resource scheduling algorithms specifically designed for the GPT-based model. Furthermore, we delve into the future directions for resource management in the GPT-based model, highlighting potential areas for further exploration and improvement. Through this study, we aim to provide valuable insights into resource management for GPT-based models deployed in clouds and promote their sustainable development for GPT-based models and applications.

cs.DC↗

arXiv2023

Learning the Unlearnable: Adversarial Augmentations Suppress Unlearnable Example Attacks

Unlearnable example attacks are data poisoning techniques that can be used to safeguard public data against unauthorized use for training deep learning models. These methods add stealthy perturbations to the original image, thereby making it difficult for deep learning models to learn from these training data effectively. Current research suggests that adversarial training can, to a certain degree, mitigate the impact of unlearnable example attacks, while common data augmentation methods are not effective against such poisons. Adversarial training, however, demands considerable computational resources and can result in non-trivial accuracy loss. In this paper, we introduce the UEraser method, which outperforms current defenses against different types of state-of-the-art unlearnable example attacks through a combination of effective data augmentation policies and loss-maximizing adversarial augmentations. In stark contrast to the current SOTA adversarial training methods, UEraser uses adversarial augmentations, which extends beyond the confines of $ \ell_p $ perturbation budget assumed by current unlearning attacks and defenses. It also helps to improve the model's generalization ability, thus protecting against accuracy loss. UEraser wipes out the unlearning effect with error-maximizing data augmentations, thus restoring trained model accuracies. Interestingly, UEraser-Lite, a fast variant without adversarial augmentations, is also highly effective in preserving clean accuracies. On challenging unlearnable CIFAR-10, CIFAR-100, SVHN, and ImageNet-subset datasets produced with various attacks, it achieves results that are comparable to those obtained during clean training. We also demonstrate its efficacy against possible adaptive attacks. Our code is open source and available to the deep learning community: https://github.com/lafeat/ueraser.

cs.LG↗

arXiv2023

Heterogeneous Domain Adaptation for IoT Intrusion Detection: A Geometric Graph Alignment Approach

Data scarcity hinders the usability of data-dependent algorithms when tackling IoT intrusion detection (IID). To address this, we utilise the data rich network intrusion detection (NID) domain to facilitate more accurate intrusion detection for IID domains. In this paper, a Geometric Graph Alignment (GGA) approach is leveraged to mask the geometric heterogeneities between domains for better intrusion knowledge transfer. Specifically, each intrusion domain is formulated as a graph where vertices and edges represent intrusion categories and category-wise interrelationships, respectively. The overall shape is preserved via a confused discriminator incapable to identify adjacency matrices between different intrusion domain graphs. A rotation avoidance mechanism and a centre point matching mechanism is used to avoid graph misalignment due to rotation and symmetry, respectively. Besides, category-wise semantic knowledge is transferred to act as vertex-level alignment. To exploit the target data, a pseudo-label election mechanism that jointly considers network prediction, geometric property and neighbourhood information is used to produce fine-grained pseudo-label assignment. Upon aligning the intrusion graphs geometrically from different granularities, the transferred intrusion knowledge can boost IID performance. Comprehensive experiments on several intrusion datasets demonstrate state-of-the-art performance of the GGA approach and validate the usefulness of GGA constituting components.

cs.CR↗

arXiv2022

Flareon: Stealthy any2any Backdoor Injection via Poisoned Augmentation

Open software supply chain attacks, once successful, can exact heavy costs in mission-critical applications. As open-source ecosystems for deep learning flourish and become increasingly universal, they present attackers previously unexplored avenues to code-inject malicious backdoors in deep neural network models. This paper proposes Flareon, a small, stealthy, seemingly harmless code modification that specifically targets the data augmentation pipeline with motion-based triggers. Flareon neither alters ground-truth labels, nor modifies the training loss objective, nor does it assume prior knowledge of the victim model architecture, training data, and training hyperparameters. Yet, it has a surprisingly large ramification on training -- models trained under Flareon learn powerful target-conditional (or "any2any") backdoors. The resulting models can exhibit high attack success rates for any target choices and better clean accuracies than backdoor attacks that not only seize greater control, but also assume more restrictive attack capabilities. We also demonstrate the effectiveness of Flareon against recent defenses. Flareon is fully open-source and available online to the deep learning community: https://github.com/lafeat/flareon.

cs.CR↗

arXiv2022

Edge-Assisted V2X Motion Planning and Power Control Under Channel Uncertainty

Edge-assisted vehicle-to-everything (V2X) motion planning is an emerging paradigm to achieve safe and efficient autonomous driving, since it leverages the global position information shared among multiple vehicles. However, due to the imperfect channel state information (CSI), the position information of vehicles may become outdated and inaccurate. Conventional methods ignoring the communication delays could severely jeopardize driving safety. To fill this gap, this paper proposes a robust V2X motion planning policy that adapts between competitive driving under a low communication delay and conservative driving under a high communication delay, and guarantees small communication delays at key waypoints via power control. This is achieved by integrating the vehicle mobility and communication delay models and solving a joint design of motion planning and power control problem via the block coordinate descent framework. Simulation results show that the proposed driving policy achieves the smallest collision ratio compared with other benchmark policies.

cs.RO↗

arXiv2022

Joint Semantic Transfer Network for IoT Intrusion Detection

In this paper, we propose a Joint Semantic Transfer Network (JSTN) towards effective intrusion detection for large-scale scarcely labelled IoT domain. As a multi-source heterogeneous domain adaptation (MS-HDA) method, the JSTN integrates a knowledge rich network intrusion (NI) domain and another small-scale IoT intrusion (II) domain as source domains, and preserves intrinsic semantic properties to assist target II domain intrusion detection. The JSTN jointly transfers the following three semantics to learn a domain-invariant and discriminative feature representation. The scenario semantic endows source NI and II domain with characteristics from each other to ease the knowledge transfer process via a confused domain discriminator and categorical distribution knowledge preservation. It also reduces the source-target discrepancy to make the shared feature space domain-invariant. Meanwhile, the weighted implicit semantic transfer boosts discriminability via a fine-grained knowledge preservation, which transfers the source categorical distribution to the target domain. The source-target divergence guides the importance weighting during knowledge preservation to reflect the degree of knowledge learning. Additionally, the hierarchical explicit semantic alignment performs centroid-level and representative-level alignment with the help of a geometric similarity-aware pseudo-label refiner, which exploits the value of unlabelled target II domain and explicitly aligns feature representations from a global and local perspective in a concentrated manner. Comprehensive experiments on various tasks verify the superiority of the JSTN against state-of-the-art comparing methods, on average a 10.3% of accuracy boost is achieved. The statistical soundness of each constituting component and the computational efficiency are also verified.

cs.CR↗