Growing data volumes present a mounting challenge to relational joins. GPUs have gained widespread adoption as database accelerators for operators such as joins due to their high instruction throughput and memory bandwidth. Most published GPU-accelerated joins are single-GPU algorithms that do not leverage modern multi-GPU platforms effectively. The few proposed multi-GPU algorithms either fail to exploit the high-speed P2P interconnects between the GPUs or to handle large out-of-core data natively. In this paper, we present a heterogeneous multi-GPU sort-merge join that overcomes both limitations. It is composed of a merge- or radix partitioning-based P2P-enabled multi-GPU sort phase, a parallel CPU-based multiway merge phase, and a hybrid join phase that combines a CPU merge path partition with a binary search-based multi-GPU join strategy. We evaluate our novel multi-GPU join on two platforms with fast NVLink- and NVSwitch-based P2P interconnects. We show that our join outperforms state-of-the-art CPU and GPU baselines regardless of the workload. It outperforms parallel CPU sort-merge and radix-hash joins by up to 15.2× and 5.5×, respectively. Compared to non-P2P-enabled multi-GPU joins, it achieves speedups of 8.7× (sort-merge) and 2.5× (hybrid-radix). We measure that our join's hybrid join phase with overlapped copy and compute operations contributes as little as 22% to its end-to-end runtime. If the input relations are pre-sorted, it is up to 14.4× faster than the hybrid-radix join. Our join scales well with the number of GPUs and benefits from data skew with as much as 12% shorter join durations.

Capturing the characteristics of real-world analytical workloads is challenging yet critical for advancing industry practices and academic research. Historically, obtaining accurate query and data characteristics has been difficult, largely because detailed workload information has often been confined to on-premises database systems. With the rise of cloud-native databases like Snowflake, it has become possible to analyze production query workloads at scale and in greater detail. Leveraging this capability, this study presents a comprehensive analysis of analytics workloads across diverse customers and industries. In particular, we investigate the query characteristics of 667 million queries issued by the most popular BI tools against Snowflake over a two-week period. Based on this dataset, this paper makes two primary contributions: first, we conduct a detailed examination of query properties, with particular attention to filters, joins, aggregations, and other previously underexplored aspects. Second, we uncover unique and practically relevant query patterns that are typically absent from standard database benchmarks.

GPUs have become a mainstream accelerator for database operations such as sorting. Most GPU sorting algorithms are single-GPU approaches. They neither harness the full computational power nor exploit the high-bandwidth P2P interconnects of modern multi-GPU platforms. The latest NVLink 2.0 and NVLink 3.0-based NVSwitch interconnects promise unparalleled multi-GPU acceleration. So far, multi-GPU sorting has only been evaluated on systems with PCIe 3.0. In this paper, we analyze serial, parallel, and bidirectional data transfer rates to, from, and between multiple GPUs on systems with PCIe 3.0/4.0, NVLink 2.0/3.0, and NVSwitch. We measure up to 35x higher parallel P2P throughput with NVLink 3.0-based NVSwitch over PCIe 3.0. To study GPU-accelerated sorting on today's hardware, we implement a P2P-based GPU-only (P2P sort) and a heterogeneous (HET sort) multi-GPU sorting algorithm and evaluate them on three modern platforms. We observe speedups over state-of-the-art parallel CPU radix sort of up to 14x for P2P sort and 9x for HET sort. On systems with fast P2P interconnects, P2P sort outperforms HET sort up to 1.65x. Finally, we show that overlapping GPU copy/compute operations does not mitigate the transfer bottleneck when sorting large out-of-core data.