• Publications
  • Influence
Anomaly Detection and Diagnosis for Container-Based Microservices with Performance Monitoring
TLDR
An anomaly detection system (ADS) is designed to detect and diagnose the anomalies in microservices by monitoring and analyzing real-time performance data of them by using machine learning models and fault injection modules integrated for training these models. Expand
High availability verification framework for OpenStack based on fault injection
The phenomenon of high availability (HA) is of vital importance in cloud architecture. This paper proposes an HA verification framework, called HAVerifier, for OpenStack, a popular open source cloudExpand
A Method of Building Fault Repository for Cloud Infrastructure
TLDR
A method of building a fault repository which can classify and record the faults that may occur in the cloud infrastructure and a domain-specific language is proposed to make the operation easier. Expand
Anomaly Detection and Diagnosis for Container-based Microservices with Performance Monitoring
TLDR
The proposed ADS consists of a monitoring module that collects the performance data of containers, a data processing module based on machine learning models and a fault injection module integrated for training these models. Expand
An approach of collecting performance anomaly dataset for NFV Infrastructure
TLDR
A method for collecting anomaly data from Infrastructure as a Service (IaaS), and constructs an anomaly database for NFV applications, and uses Kubernetes to build a distributed environment to accelerate the occurrence of anomalies. Expand
An Approach of Collecting Performance Anomaly Dataset for NFV Infrastructure
TLDR
This paper designs a method for collecting anomaly data from Infrastructure as a Service (IaaS), and constructs an anomaly database for NFV applications, and uses Kubernetes to build a distributed environment. Expand