DynIMS: A Dynamic Memory Controller for In-memory Storage on HPC Systems

TL;DR

DynIMS is a dynamic memory controller that adaptively manages in-memory storage in HPC systems, significantly improving performance by optimizing memory allocation for compute and data-intensive tasks.

Contribution

This paper introduces DynIMS, a novel feedback-based control system that dynamically adjusts in-memory storage capacity based on real-time memory demands in HPC environments.

Findings

Up to 5X performance improvement with DynIMS

Effective online inference of memory demands

Enhanced resource utilization in HPC systems

Abstract

In order to boost the performance of data-intensive computing on HPC systems, in-memory computing frameworks, such as Apache Spark and Flink, use local DRAM for data storage. Optimizing the memory allocation to data storage is critical to delivering performance to traditional HPC compute jobs and throughput to data-intensive applications sharing the HPC resources. Current practices that statically configure in-memory storage may leave inadequate space for compute jobs or lose the opportunity to utilize more available space for data-intensive applications. In this paper, we explore techniques to dynamically adjust in-memory storage and make the right amount of space for compute jobs. We have developed a dynamic memory controller, DynIMS, which infers memory demands of compute tasks online and employs a feedback-based control model to adapt the capacity of in-memory storage. We test DynIMS using mixed HPCC and Spark workloads on a HPC cluster. Experimental results show that DynIMS can achieve up to 5X performance improvement compared to systems with static memory allocations.