Speaker "Seema Nagar" Details Back



Alert noise reduction for SREs


The robustness and availability of cloud services are becoming increasingly important as more applications migrate to the cloud. The operations landscape today is more complex, than ever. Site reliability engineers (SREs) are expected to handle more incidents than ever before with shorter service-level agreements (SLAs). By exploiting log, tracing, metric, and network data, Artificial Intelligence for IT Operations (AIOps) enables detection of faults and anomalous issues of services. A wide variety of anomaly detection techniques have been incorporated in various AIOps platforms (e.g. PCA and autoencoder), but they all suffer from false positives. In this paper, we propose an unsupervised approach for \textit{persistent} anomaly detection on top of the traditional anomaly detection approaches, with the goal of reducing false positives and providing more trustworthy alerting signals. We test our method on both simulated and real-world datasets. Our technique reduces false positive anomalies by at least 28%, resulting in more reliable and trustworthy notifications.


Seema Nagar is staff research scientist at IBM Research India. She has been researching the field of computer science for the last thirteen years. She has over 40 publications in eminent conferences and journals and more than 150 patents filed. She has been named a master inventor for two consecutive terms. She has also been part of the reviewer committee of many eminent conferences, such as IJCAI, NAACL, and ACL. She has earned several Research Awards in IBM Research India, including three Outstanding Technical Achievement Awards for her work on social network analysis and trustworthy AI. She actively mentors several researchers/students at the research lab. She obtained her B. Tech. and M. Tech. in Computer Science and Engineering from IIT, Guwahati in 2007, and IIT, Delhi in 2011 respectively. Currently, she is also pursuing a part-time PhD from IIIT, Guwahati