Aggregating Funnels for Faster Fetch&Add and Queues

TL;DR

The paper introduces Aggregating Funnels, a novel algorithm that reduces contention in concurrent fetch-and-add operations by batching and distributing them across multiple memory locations, significantly improving throughput and scalability.

Contribution

It proposes a new batching algorithm, Aggregating Funnels, that enhances performance and scalability of concurrent fetch-and-add operations and queues.

Findings

Achieves higher throughput than previous combining techniques.

Substantially more scalable than hardware fetch-and-add on a single location.

Eliminates bottlenecks in state-of-the-art concurrent queues.

Abstract

Many concurrent algorithms require processes to perform fetch-and-add operations on a single memory location, which can be a hot spot of contention. We present a novel algorithm called Aggregating Funnels that reduces this contention by spreading the fetch-and-add operations across multiple memory locations. It aggregates fetch-and-add operations into batches so that the batch can be performed by a single hardware fetch-and-add instruction on one location and all operations in the batch can efficiently compute their results by performing a fetch-and-add instruction on a different location. We show experimentally that this approach achieves higher throughput than previous combining techniques, such as Combining Funnels, and is substantially more scalable than applying hardware fetch-and-add instructions on a single memory location. We show that replacing the fetch-and-add instructions in the fastest state-of-the-art concurrent queue by our Aggregating Funnels eliminates a bottleneck and greatly improves the queue's overall throughput.