TY - GEN
T1 - B-Meg
T2 - 13th Annual ACM/SPEC International Conference on Performance Engineering, ICPE 2022
AU - Somashekar, Gagan
AU - Dutt, Anurag
AU - Vaddavalli, Rohith
AU - Varanasi, Sai Bhargav
AU - Gandhi, Anshul
N1 - Publisher Copyright:
© 2022 ACM.
PY - 2022/7/14
Y1 - 2022/7/14
N2 - The microservices architecture enables independent development and maintenance of application components through its fine-grained and modular design. This has enabled rapid adoption of microservices architecture to build latency-sensitive online applications. In such online applications, it is critical to detect and mitigate sources of performance degradation (bottlenecks). However, the modular design of microservices architecture leads to a large graph of interacting microservices whose influence on each other is non-trivial. In this preliminary work, we explore the effectiveness of Graph Neural Network models in detecting bottlenecks. Preliminary analysis shows that our framework, B-MEG, produces promising results, especially for applications with complex call graphs. B-MEG shows up to 15% and 14% improvements in accuracy and precision, respectively, and close to 10× increase in recall for detecting bottlenecks compared to the technique used in existing work for bottleneck detection in microservices.
AB - The microservices architecture enables independent development and maintenance of application components through its fine-grained and modular design. This has enabled rapid adoption of microservices architecture to build latency-sensitive online applications. In such online applications, it is critical to detect and mitigate sources of performance degradation (bottlenecks). However, the modular design of microservices architecture leads to a large graph of interacting microservices whose influence on each other is non-trivial. In this preliminary work, we explore the effectiveness of Graph Neural Network models in detecting bottlenecks. Preliminary analysis shows that our framework, B-MEG, produces promising results, especially for applications with complex call graphs. B-MEG shows up to 15% and 14% improvements in accuracy and precision, respectively, and close to 10× increase in recall for detecting bottlenecks compared to the technique used in existing work for bottleneck detection in microservices.
KW - anomaly detection
KW - bottleneck detection
KW - dataset
KW - graph neural networks
KW - microservices
UR - https://www.scopus.com/pages/publications/85136030379
U2 - 10.1145/3491204.3527494
DO - 10.1145/3491204.3527494
M3 - Conference contribution
AN - SCOPUS:85136030379
T3 - ICPE 2022 - Companion of the 2022 ACM/SPEC International Conference on Performance Engineering
SP - 7
EP - 11
BT - ICPE 2022 - Companion of the 2022 ACM/SPEC International Conference on Performance Engineering
PB - Association for Computing Machinery, Inc
Y2 - 9 April 2022 through 13 April 2022
ER -