Write a Blog >>
ASE 2021
Sun 14 - Sat 20 November 2021 Australia
Wed 17 Nov 2021 22:20 - 22:40 at Koala - Performance Chair(s): Ming Wen

Service reliability is one of the key challenges that cloud providers have to deal with. In cloud systems, unplanned service failures may cause severe cascading impacts on their dependent services, deteriorating customer satisfaction. Predicting the cascading impacts accurately and efficiently is critical to the operation and maintenance of cloud systems. Existing approaches identify whether one service depends on another via distributed tracing but no prior work focused on discriminating the intensity of the dependency between cloud services. In this paper, we empirically study the outages and the procedure for failure diagnosis in two cloud providers to motivate the definition of the intensity of dependency. Then we propose AIM, the first approach to predict the intensity of dependencies between cloud microservices. AIM first generates a set of candidate dependency pairs from the spans. AIM then represents the status of each cloud service with a multivariate time series aggregated from the spans. With the representation of services, AIM calculates the similarities between the statuses of the caller and callee of each candidate pair. Finally, AIM aggregates the similarities to produce a unified value as the intensity of the dependency. We evaluate AIM on the data collected from an open-source microservice benchmark and a cloud system in production. The experimental results show that AIM can efficiently and accurately predict the intensity of dependencies. We further demonstrate the usefulness of our method in a large-scale cloud system. We plan to release both datasets to facilitate future studies.

Wed 17 Nov

Displayed time zone: Hobart change

22:00 - 23:00
PerformanceResearch Papers / Journal-first Papers / Tool Demonstrations at Koala
Chair(s): Ming Wen Huazhong University of Science and Technology
22:00
20m
Talk
"What makes my queries slow?": Subgroup Discovery for SQL Workload Analysis
Research Papers
Youcef Remil Infologic, INSA Lyon, Anes Bendimerad Infologic, Romain Mathonat Infologic, Philippe Chaleat Infologic, Mehdi Kaytoue INFOLOGIC
22:20
20m
Talk
AID: Efficient Prediction of Aggregated Intensity of Dependency in Large-scale Cloud Systems
Research Papers
Tianyi Yang The Chinese University of Hong Kong, Jiacheng Shen The Chinese University of Hong Kong, Yuxin Su The Chinese University of Hong Kong, Xiao Ling Huawei Technologies, Yongqiang Yang Huawei Technologies, Michael Lyu The Chinese University of Hong Kong
22:40
10m
Talk
Assessment of Off-the-Shelf SE-specific Sentiment Analysis Tools: An Extended Replication Study
Journal-first Papers
Nicole Novielli University of Bari, Fabio Calefato University of Bari, Filippo Lanubile University of Bari, Alexander Serebrenik Eindhoven University of Technology
22:50
5m
Talk
EvoMe: A Software Evolution Management Engine Based on Differential Factbase
Tool Demonstrations
Xiuheng Wu Nanyang Technological University, Mengyang Li Nanyang Technological University, Yi Li Nanyang Technological University
Pre-print
22:55
5m
Talk
RefactorInsight: Enhancing IDE Representation of Changes in Git with Refactorings Information
Tool Demonstrations
Zarina Kurbatova JetBrains Research, Vladimir Kovalenko JetBrains Research, Ioana Savu Delft University of Technology, Bob Brockbernd Delft University of Technology, Dan Andreescu Delft University of Technology, Matei Anton Delft University of Technology, Roman Venediktov Higher School of Economics, Elena Tikhomirova JetBrains Research, Timofey Bryksin JetBrains Research; HSE University
Pre-print