English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Near-Data-Processing (NDP) architectures present a promising way to alleviatedata movement costs and can provide significant performance and energy benefitsto parallel applications. Typically, NDP architectures support several NDPunits, each including multiple simple cores placed close to memory. To fullyleverage the benefits of NDP and achieve high performance for parallelworkloads, efficient synchronization among the NDP cores of a system isnecessary. However, supporting synchronization in many NDP systems ischallenging because they lack shared caches and hardware cache coherencesupport, which are commonly used for synchronization in multicore systems, andcommunication across different NDP units can be expensive.  This paper comprehensively examines the synchronization problem in NDPsystems, and proposes SynCron, an end-to-end synchronization solution for NDPsystems. SynCron adds low-cost hardware support near memory for synchronizationacceleration, and avoids the need for hardware cache coherence support. SynCronhas three components: 1) a specialized cache memory structure to avoid memoryaccesses for synchronization and minimize latency overheads, 2) a hierarchicalmessage-passing communication protocol to minimize expensive communicationacross NDP units of the system, and 3) a hardware-only overflow managementscheme to avoid performance degradation when hardware resources forsynchronization tracking are exceeded.  We evaluate SynCron using a variety of parallel workloads, covering variouscontention scenarios. SynCron improves performance by 1.27$\times$ on average(up to 1.78$\times$) under high-contention scenarios, and by 1.35$\times$ onaverage (up to 2.29$\times$) under low-contention real applications, comparedto state-of-the-art approaches. SynCron reduces system energy consumption by2.08$\times$ on average (up to 4.25$\times$).

SynCron: Efficient Synchronization Support for Near-Data-Processing Architectures