Proceedings of the 29th annual ACM/IEEE international symposium on Microarchitecture
Wrappers for feature subset selection
Artificial Intelligence - Special issue on relevance
Information Retrieval
Fast Algorithms for Mining Association Rules in Large Databases
VLDB '94 Proceedings of the 20th International Conference on Very Large Data Bases
An extensive empirical study of feature selection metrics for text classification
The Journal of Machine Learning Research
Magpie: online modelling and performance-aware systems
HOTOS'03 Proceedings of the 9th conference on Hot Topics in Operating Systems - Volume 9
A Branch and Bound Algorithm for Feature Subset Selection
IEEE Transactions on Computers
Experiences with tracing causality in networked services
INM/WREN'10 Proceedings of the 2010 internet network management conference on Research on enterprise networking
Diagnosing performance changes by comparing request flows
Proceedings of the 8th USENIX conference on Networked systems design and implementation
X-trace: a pervasive network tracing framework
NSDI'07 Proceedings of the 4th USENIX conference on Networked systems design & implementation
Modeling the parallel execution of black-box services
HotCloud'11 Proceedings of the 3rd USENIX conference on Hot topics in cloud computing
Evaluating web user perceived latency using server side measurements
Computer Communications
| Hi-index | 0.00 |
Data centers run many services that impact millions of users daily. In reality, the latency of each service varies from one request to another. Existing tools allow to monitor services for performance glitches or service disruptions, but typically they do not help understanding the variations in latency. We propose a general framework for understanding performance of arbitrary black box services. We consider a stream of requests to a given service with their monitored attributes, as well as latencies of serving each request. We propose what we call the multi-dimensional f-measure, that helps for a given interval to identify the subset of monitored attributes that explains it. We design algorithms that use this measure not only for a fixed latency interval, but also to explain the entire range of latencies of the service by segmenting it into smaller intervals. We perform a detailed experimental study with synthetic data, as well as real data from a large search engine. Our experiments show that our methods automatically identify significant latency intervals together with request attributes that explain them, and are robust.