The Art of Efficient In-memory Query Processing on NUMA Systems: a Systematic Approach

Data analytics systems commonly utilize in-memory query processing techniques to achieve better throughput and lower latency. Modern computers increasingly rely on Non-Uniform Memory Access (NUMA) architectures to achieve scalability. A key drawback of NUMA architectures is that many existing software solutions are not aware of the underlying NUMA topology and thus do not take full advantage of the hardware. Modern operating systems are designed to provide basic support for NUMA systems. However, default system configurations are typically sub-optimal for large data analytics applications. Additionally, rewriting the application from the ground up is not always feasible.

In this work, we evaluate a variety of strategies that aim to accelerate memory-intensive data analytics workloads on NUMA systems. Our findings indicate that the operating system default configurations can be detrimental to query performance. We analyze the impact of different memory allocators, memory placement strategies, thread placement, and kernel-level load balancing and memory management mechanisms. With extensive experimental evaluation, we demonstrate that the methodical application of these techniques can be used to obtain significant speedups in four commonplace in-memory query processing tasks, on three different hardware architectures. Furthermore, we show that these strategies can improve the performance of five popular database systems running a TPC-H workload. Lastly, we summarize our findings in a decision flowchart for practitioners.