In cloud computing, how to reasonably allocate computing resources for batch jobs to ensure the load balance of dynamic clusters and meet user requests is an important and challenging task. Most existing studies are based on deep Q network, which utilizes neural networks to estimate the expected value of cumulative return in the scheduling process. The value-based DQN algorithms ignore the complete information contained in the value distribution and lack strong adaptability to time-varying batch jobs and dynamic cluster resources. Therefore, to capture the inherent stochasticity of the scheduling process caused by environmental stochasticity, we utilize Distributional Reinforcement Learning to model the value distribution of the cumulative return. Specifically, we formalize the load balancing scheduling as a multi-objective optimization problem and construct a Distributional Reinforcement Learning model. Then we introduce quantile regression to learn the value distribution of the cumulative return during scheduling and propose a dynamic load balancing scheduling algorithm based on Distributional Reinforcement Learning. In addition, we develop a cluster environment for real-time processing of batch jobs to simulate the arrival of batch jobs and train the Distributional Reinforcement Learning-based scheduling agent. We conduct empirical experiments and detailed analysis by using the real Alibaba Cluster cluster traces v2018 and v2020. The results show that compared to the baseline algorithms, the proposed algorithm performs better in terms of cluster load balancing, success rate of instance creation and average completion time of the tasks. The experimental results on different trace datasets also indicate that the propsoed algorithm exhibits excellent scalability.

Nowadays, microservice architecture has become mainstream way of cloud applications delivery. Distributed tracing is crucial to preserve the observability of microservice systems. However, existing trace representation approaches only concentrate on operations, relationships and metrics related to service invocations. They ignore service events that denotes meaningful, singular point in time during the service's duration. In this paper, we propose iTCRL, a novel trace contrastive representation learning approach based on causal intervention. This approach first constructs a unified graph representation for each trace to describe the runtime status of service events in traces and the complex relationships between them. Then, Causal-intervention-based Trace Contrastive Learning is proposed, which learns trace representations from causal perspective based on the unified graph representations of traces. It uses causal intervention to generate contrastive views, heterogeneous graph neural network-based trace encoder to learn trace representations, and direct causal effect to guide the training of trace encoder. Experimental results on three datasets show that iTCRL outperforms all baselines in terms of trace classification, trace anomaly detection, trace sampling and noise robustness, and also validate the contribution of Causal-intervention-based Trace Contrastive Learning.

Logs that record system abnormal states (anomaly logs) can be regarded as outliers, and the k-Nearest Neighbor (kNN) algorithm has relatively high accuracy in outlier detection methods. Therefore, we use the kNN algorithm to detect anomalies in the log data. However, there are some problems when using the kNN algorithm to detect anomalies, three of which are: excessive vector dimension leads to inefficient kNN algorithm, unlabeled log data cannot support the kNN algorithm, and the imbalance of the number of log data distorts the classification decision of kNN algorithm. In order to solve these three problems, we propose an efficient log anomaly detection method based on an improved kNN algorithm with an automatically labeled sample set. This method first proposes a log parsing method based on N-gram and frequent pattern mining (FPM) method, which reduces the dimension of the log vector converted with Term frequency.Inverse Document Frequency (TF-IDF) technology. Then we use clustering and self-training method to get labeled log data sample set from historical logs automatically. Finally, we improve the kNN algorithm using average weighting technology, which improves the accuracy of the kNN algorithm on unbalanced samples. The method in this article is validated on six log datasets with different types.

Vehicular crowdsourcing networks (VCNs) enable vehicles to provide or obtain traffic-related services in a costefficient and flexible manner. Therefore, it is crucial to provide trusted management in VCNs for high reliability towards both service producers and consumers. However, most recent VCN platforms rely on a third party to manage crowdsourcing services which might be not fully trusted by users. For the issue, this paper proposes a blockchain-based trust management scheme for VCNs to provide a decentralized and trusted service management. A comprehensive trust evaluation model (TEM) is designed to quantify the trust degree of each vehicular node, and a vehicle-trust blockchain framework called VTchain is proposed to preserve the trust values of nodes while guaranteeing transparency and trustworthiness. Particularly, we leverage a trusted execution environment (TEE) to provide secure trust evaluation to tackle possible untrusted road-side units. In addition, we introduce TEM-based Proof of Trust to support blockchain maintenance, which works together with an efficient consensus algorithm Zyzzyva for improved scalability. Finally, extensive experiments are conducted by developing a testbed deployed on cloud servers for measurements.

Urban function zoning is of great importance for urban structure optimization, urban resource allocation, and urban development planning. Since citizens usually act as a network of motion sensors of the city, their activities could reflect the environment around them. We considered taking advantage of VGI data to classify urban function zones. In this paper, we proposed a framework for automated urban function zoning which is based on VGI geo-tagged photos and OpenStreetMap (OSM) data. Through combining the high-level image features of geo-tagged photos with the road network data, we obtained the functional zoning map of the study area. The experiment result shows the effectiveness of the framework we proposed.

In this paper, a class of closed queueing networks is modelled in the Markovian process algebra PEPA and solved using the classical mean value analysis. This approach is attractive as it negates the need to derive the entire state space, and so certain metrics from large models can be obtained with little computational effort. The class of models considered includes models that are not obviously classical closed queueing models. This paper extends previous results by deriving exact solutions for models with multiple copies of queue stations. Due to the cost of deriving an exact solution, a simplifying approximation is derived, which is shown to be highly accurate when the population is large. The approach is illustrated with three examples.

In this paper we study the overhead introduced by secure functions in considering two models of non-repudiation protocols. The models are specified using the Markovian process algebra PEPA. The basic model suffers from the well known state space explosion problem when tackled using Markov chain analysis. Following previous study of performance modelling on security protocols, mean value analysis and fluid flow approximation based on ordinary differential equations (ODEs) have been chosen as efficient analysis techniques. Mean value analysis is an efficient exact method for deriving a limited set of metrics for large numbers of clients involved in the protocols. Fluid flow approximations can be adopted to solve the system with extremely large populations and potentially derive a wider range of metrics. The models are analyzed numerically and results derived from mean value analysis are compared with the ODE solution.