Practical Resource Usage Prediction Method for Large Memory Jobs in HPC Clusters

Abstract

Users in high performance computing (HPC) clusters normally face challenges to specify accurate resource estimates for running their applications as batch jobs. Prediction is a common way to alleviate this complexity by using historical job records of previous runs to estimate resource usage for new coming jobs. Most of existing resource prediction methods directly build a single model to consider all of the jobs in clusters. However, people in production usage tend to only focus on the resource usage of jobs with certain patterns, e.g. jobs with large memory consumption. This paper proposes a practical resource prediction method for large memory jobs. The proposed method first tries to predict whether a job tends to use large memory size, and then predicts the final memory usage using a model which is trained by only historical large memory jobs. Using several real-world job traces collected from large production clusters of IBM Spectrum LSF customer sites, the evaluation results show that the average prediction errors can be reduced up to 40% for nearly 90% of large memory jobs. Meanwhile, the model training cost can be reduced over 30% for the evaluated job traces.

Preemptive Parallel Job Scheduling for Heterogeneous Systems Supporting Urgent Computing

“…First, most of the NPB applications have a process size smaller than 1.25 gigabytes. Second, as shown in the LANL workload [54] and three real-world workloads of a related study [55], each of the majority of jobs, not a process, uses no more than 1 gigabyte of memory. In Table 1 of [55], the total percentage of the small memory jobs is much higher than that of the large memory jobs.…”

“…Second, as shown in the LANL workload [54] and three real-world workloads of a related study [55], each of the majority of jobs, not a process, uses no more than 1 gigabyte of memory. In Table 1 of [55], the total percentage of the small memory jobs is much higher than that of the large memory jobs. Third, the results of a previous study [56] show that 10 of the 13 production HPC applications under investigation have process sizes in the range of hundreds of megabytes.…”