The user experience for networked applications is becoming a key benchmark for customers and network providers. Perceived u ser experience is largely determined by the frequency, duratio n and severity of network events that impact a service. While toda y’s networks implement sophisticated infrastructure that issues alarms for most failures, there remains a class of silent outages (e.g. , caused by configuration errors) that are not detected. Further, exist ing alarms provide little information to help operators understand th e impact of network events on services. Attempts to address this thro ugh infrastructure that monitors end-to-end performance for c ustomers have been hampered by the cost of deployment and by the volume of data generated by these solutions. We present an alternative approach that pushes monitoring t o applications on end systems and uses their collective view t o detect network events and their impact on services - an appro ach we call Crowdsourcing Event Monitoring (CEM). This paper presents a general framework for CEM systems and demonstrat es its effectiveness for a P2P application using a large datase t gathered from BitTorrent users and confirmed network events from two ISPs. We discuss how we designed and deployed a prototype CEM implementation as an extension to BitTorrent. This syst em performs online service-level network event detection thr ough passive monitoring and correlation of performance in end-u sers’ applications. Categories and Subject Descriptors C.2.3 [ Network Operations ]: Network monitoring C.2.4 Distributed Systems Distributed Applications General Terms Measurement, Performance, Reliability