Valetto Giuseppe author Kaiser Gail E. author 2002 2009-02-24T20:31:46Z text application/pdf We have developed an infrastructure for end-to-end run-time monitoring, behavior / performance analysis, and dynamic adaptation of distributed software applications. This feedback-loop infrastructure is primarily targeted to pre-existing systems and thus operates outside the application itself without making assumptions about the target system's internal communication/computation mechanisms, implementation language/framework, availability of source code, etc. This paper assumes the existence of the monitoring and analysis components, presented elsewhere, and focuses on the mechanisms used to control and coordinate possibly complex repairs/reconfigurations to the target system. These mechanisms require lower-level actuators or effectors somehow attached to the target system, so we briefly sketch one such facility (elaborated elsewhere). The core of the paper is the model, architecture, and implementation of Workflakes, the decentralized process engine we use to tailor, control, coordinate, respond to contingencies, etc. regarding a cohort of such actuators. We have validated our approach and the Workflakes prototype in several case studies, related to different application domains. Due to space restrictions we concentrate primarily on one case study, elaborate with some details a second, and only sketch others. English 151 http://hdl.handle.net/10022/AC:P:29293 NNC NNC 2009-06-24 2009-02-24 23:02:18 urn:hdl:10022/AC:P:29293 eng Made available in DSpace on 2009-02-24T20:31:46Z (GMT). No. of bitstreams: 1151.pdf: 129843 bytes, checksum: da008f017e48fe6a5d8750c112b7f7df (MD5) Previous issue date: 2002-09-14