Continuity in Task Execution Sample Clauses

Continuity in Task Execution. ‌ The problem of continuous or pseudo-continuous task execution arises in diverse areas of life (e.g. plants, robots, air pollution monitoring). Some research seeks to find an optimal schedule for continuous processes (e.g. continuous mixing of feedstock) to utilise some resources (e.g. equipment) on plants [34], while other research addresses a dynamic area partitioning among a group of robots in order to monitor this area continuously in respect of some events of interest (e.g. gas leaks) [35]. The work of Xxxx et al. [6] proposes a Grid-based Adaptive Execution on Streams (GATES) to process distributed data streams in a real-time manner. For instance, this Grid system can be used to process tasks which can predict large earthquakes based on ground movements. Xxxx et al. also argue that a Grid environment is well-suited for the effective processing of data streams (i.e. continuous tasks). That is, the large-scale and distributed Grid might be able to allocate computing resources which are geographically close to the source of data that might reduce communication cost. XXXXX aims to dynamically adjust the amount of arriving data with the amount of available computational, storage and communication resources to reach the best accuracy of results in real-time. Some other systems which process data streams, such as DGClust [8], StarGlobe [36], Calder [37] propose mechanisms to decrease computational and communication costs of processing data streams. For example, StarGlobe applies filtering and parallel processing of data streams to decrease the level of resource load. In order to process continuous data stream, continuous queries are discussed in many approaches [2, 38–41], and it is defined as the type of query issued once for a particular data-type and then runs continuously, updating a client with new results without being issued repeatedly. In other words, a continuous query is repeated over time to process continuous data streams. Babu and Widom [38] state that processing of continuous data streams may require unbounded storage for the data traces. Therefore, they propose an architecture of continuous query processing, where an answer to a query can be sent to a permanent or temporary storage and it can later be deleted from a temporary storage. Newly arrived data can be also saved for some time and deleted later. In this way, the authors remove unnecessary data e.g., some outdated data, to avoid storage excess. The idea of discriminative data storing ...