Scalable Concurrent B-Trees Using Multi-Version Memory

Wang, Paul; Weihl, William E.

doi:10.1006/jpdc.1996.0003

Cited by 2 publications

(1 citation statement)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the writer has finished its update(s), it publishes the new version such that subsequent readers will operate on the new version. Wang and Weihl describe a software caching algorithm used in an implementation of concurrent B-trees that allows multiple versions of memory such that local stale versions of the B-tree can be read without incurring a cache miss, saving the latency of fetching the new version [30]. Their implementation improves performance over 300% for a highly contended B-tree microbenchmark.…”

Section: Software Systemsmentioning

confidence: 99%

Edge chasing delayed consistency

Cain

Lipasti

2012

Proceedings of the 2012 ACM Workshop on Relaxing Synchronization for Multicore and Manycore Scalability

View full text Add to dashboard Cite

In shared memory multiprocessors utilizing invalidationbased coherence protocols, cache misses caused by interprocessor communication are a dominant source of processor stall cycles for many applications. We explore a novel coherence protocol implementation called edge-chasing delayed consistency (ECDC) that mitigates some of the performance degradation caused by this class of misses. Edgechasing delayed consistency allows a processor to nonspeculatively continue reading a cache line after receiving an invalidation from another core, without changing the consistency model offered to programmers. While the idea of using stale data for as long as possible is enticing, our study shows that the benefits of such delay are small, and that the majority of these delayed invalidation benefits come from mitigating the false sharing problem, rather than any tolerance of races or an application's ability to consume stale data in a productive manner.

show abstract

Section: Software Systemsmentioning

confidence: 99%