Data movement patterns slow down DNA mapping tools on modern CPUs
In-depth data movement characterization of DNA read mapping algorithms
Hardware Architecture
Summary
DNA read mapping tools help scientists compare DNA sequences by matching small pieces of DNA to a reference genome. These tools can be slow and use a lot of computer memory, which limits how fast genetic analysis can run. The authors studied several popular DNA mapping programs to see how they use the computer’s memory and processor cache. They found that some programs don’t use computer memory efficiently, causing delays, while others run faster or slower because of how well they use multiple CPU cores. This analysis helps identify where improvements or new hardware could speed up DNA analysis.
What this means in practice
- •For bioinformatics developers: Optimize DNA read mapping software by focusing on memory use and cache efficiency to improve speed on typical CPUs.
- •For hardware designers: Design processor-in-memory architectures targeting DNA mapping algorithms that do not fully exploit current CPU cache hierarchies.
Authors
Berkan Şahin, Ecem İlgün, Klea Zambaku, Can Alkan
Abstract
Read mapping algorithms represent a critical bottleneck in genomic analysis pipelines due to the resources required; a problem that is becoming more pronounced as sequencing technologies advance and datasets expand. We present an extensive workload characterization of leading read mapping algorithms across diverse sequencing technologies to identify hardware acceleration and optimization opportunities. We analyzed the following implementations: MashMap, wfmash, and Minimap2 for whole-genome alignment (WGA); NGMLR, Minimap2, and LRA for long reads; and Bowtie2, BWA-MEM, StrobeAlign, and minibwa for short reads. We used the Intel VTune Profiler for detailed memory analysis to identify memory- bound functions, cache usage patterns, and data transfer bottlenecks. We further analyzed a selected subset of alignment tools using the DAMOV-SIM simulator to obtain a more granular view of data movement characteristics for two different cache structures. Based on our results, Minimap2 emerges as the fastest alignment tool for long read scenarios, but falls behind in whole-genome alignment due to poor multithreading. Across all three categories, some programs (e.g., LRA, BWA-MEM) are highly memory-bound. More in-depth analysis via DAMOV-SIM revealed that some memory-bound tools, in particular, do not fully utilize the multi-level cache hierarchy in modern multi-core CPU architectures. Our results show that while some tools require algorithmic redesign for performance gains, some others can benefit from running on PIM architectures.