Survey Study on Scalable Scheduling of Updates in Streaming Datawarehouses

Computer Engineering

This survey deals with the update scheduling problem in streaming data warehouse, The proposed multitrack algorithm solves the problem of updating streaming data warehouse by scheduling the jobs where jobs corresponds to updates for tables as well as views. Those jobs load new data into tables, aim is to minimize the data staleness over time. For scaling purpose we use partitioning strategies. The partitioning strategies such as EDF partitioning and some part of proportional partitioning, are combined in multitrack algorithm. We calculate track utilization, if track utilization is higher then updates are done faster in data warehouse. Another paper focuses on the update scheduling problem, when the views are materialized, find the best order to refresh them to maximize the quality of data (QoD) in presence of incessant updates. The group-EDF (gEDF) algorithm is based on dynamic grouping of tasks with deadlines. Deadlines are very close to each other, SJF algorithm is used to schedule task within a group. Hence total execution time is minimized. Using Round Robin each job is assigned a time interval, called its quantum, which is allowed to run. If the job is still running at the end of quantum, the CPU is preempted and given to another job. If the job has blocked or finished before the quantum has elapsed, the CPU switching is done. Of course, round robin is easy to implement.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.