Grant: Discovery Grants Program - Individual

Quality-of-service (QoS)-enabled publish/subscribe (pub/sub) middleware provides powerful support for scalable data dissemination. It is hard, however, to maintain key quality of service properties (such as reliability and latency) in flexible environments (e.g., cloud computing infrastructure) and operating environments that change over time for distributed real-time and embedded (DRE) systems (such as disaster relief operations or power grids). Managing quality of service manually is often not feasible in these flexible and dynamic environments due to slow response times, the complexity of managing multiple interrelated quality of service settings, and the scale of the systems being managed. For certain domains, DRE systems must be able to reflect on the conditions of their environment and adapt accordingly in a bounded amount of time. This research focuses on autonomic adaptation of QoS-enabled pub/sub DRE middleware. In particular, autonomic adaptation is needed for both configuration adaptation within flexible environments where the operating environment is unknown until runtime and ongoing modifications within dynamic environments where the operating environment changes at runtime. This research addresses key areas in adapting pub/sub middleware to support QoS in flexible and dynamic environments including (1) creating new composite metrics and leveraging existing composite metrics as needed to quantitatively evaluate multiple QoS concerns (e.g., reliability and latency), (2) creating new QoS policies and precisely defining the semantics of combining existing policies that allow an application developer to express needed composite metrics, (3) creating a taxonomy of transport protocol properties and a mapping of QoS policies (including newly created policies) to transport protocol properties, (4) configuring QoS-enabled pub/sub middleware autonomically for environments where the environment configuration is unknown beforehand (e.g., cloud environments) and dynamically adapting QoS mechanisms in response to changes in the environment that affect QoS (e.g., percent network loss, # of data senders and receivers).