Salamander: a Holistic Scheduling of MapReduce Jobs on Ephemeral Cloud Resources

Abstract

Most cloud data centers are over-provisioned and underutilized, primarily to handle peak loads and sudden failures. This has motivated many researchers to reclaim the unused resources, which are by nature ephemeral, to run data-intensive applications at a lower cost. Hadoop MapReduce is one of those applications. However, it was designed on the assumption that resources are available as long as users pay for the service. In order to make it possible for Hadoop to run on unused (ephemeral) resources, we have designed a heterogeneity and volatility-aware holistic scheduler consisting of three different components: (1) A MapReduce task and job scheduler that relies on a global vision of resource utilization predictions, (2) a scheduler-based data placement strategy that improves the data locality, and (3) a reactive QoS controller that ensures customers' service-level agreement (SLA) and minimizes interference between co-located workloads. Our framework makes it possible to take advantage of ephemeral resources efficiently. Indeed, for a given set of jobs, it reduces the overall execution time by up to 47.6% and an average of 18.7% as compared to state-of-the-art strategies.

Handling Non-Local Executions to Improve MapReduce Performance Using Ant Colony Optimization

“…We simulated MapReduce task scheduler (Hadoop 2.7.0) to evaluate the proposed methodology on Ubuntu server with 12-core CPU (hyper-threaded), 64 GB memory, storage 4 x 1 TB HDD, and disk bandwidth rate 100 MB maximum. We compared IDLACO with classical fair scheduler and Holistic scheduler [16] based on parameters, such as the This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/ This article has been accepted for publication in a future issue of this journal, but has not been fully edited.…”

“…The proposed approach largely relies on data locality to minimize the job latency with frequent replications of data blocks on demand. A holistic scheduler was designed by Mohamed Handaoui et al in [16] to improve resource utilization and job latency. It consists of three components: resource utilization prediction, determining data local executions, minimizing interferences from co-locating workloads.…”

ReLeaSER: A Reinforcement Learning Strategy for Optimizing Utilization Of Ephemeral Cloud Resources

“…These predictions are used to reclaim unused resources in order to be allocated to customers. The prediction shows good general accuracy and was used in other studies for scheduling big data applications [8], [9]. We followed 2 steps in this section: 1) Using the predictions alongside the in-production traces, we analyzed the prediction errors on different granularities: datacenter, host, resource metric.…”

“…The safety margin may be a static value, that is a fixed proportion of resources applied all the time, for all hosts and resource metrics. This strategy was used in Cuckoo [8] and Salamander [9] where fixed proportions were empirically tested to select the best one. Although this strategy does decrease potential SLA violations, a substantial amount of resources remained unused due to resource usage overestimations [9].…”

“…This strategy was used in Cuckoo [8] and Salamander [9] where fixed proportions were empirically tested to select the best one. Although this strategy does decrease potential SLA violations, a substantial amount of resources remained unused due to resource usage overestimations [9]. Moreover, the prediction accuracy of the CPU proved to be lower than that of the RAM [2], which means that the safety margin should be customized for each resource metric.…”