FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2321262893> ?p ?o ?g. }

• W2321262893 abstract "SIMD organizations amortize the area and power of fetch, decode, and issue logic across multiple processing units in order to maximize throughput for a given area and power budget. However, throughput is reduced when a set of threads operating in lockstep (a warp) are stalled due to long latency memory accesses. The resulting idle cycles are extremely costly. Multi-threading can hide latencies by interleaving the execution of multiple warps, but deep multi-threading using many warps dramatically increases the cost of the register files (multi-threading depth SIMD width), and cache contention can make performance worse. Instead, intra-warp latency hiding should first be exploited. This allows threads that are ready but stalled by SIMD restrictions to use these idle cycles and reduces the need for multi-threading among warps. This paper introduces dynamic warp subdivision (DWS), which allows a single warp to occupy more than one slot in the scheduler without requiring extra register file space. Independent scheduling entities allow divergent branch paths to interleave their execution, and allow threads that hit to run ahead. The result is improved latency hiding and memory level parallelism (MLP). We evaluate the technique on a coherent cache hierarchy with private L1 caches and a shared L2 cache. With an area overhead of less than 1%, experiments with eight data-parallel benchmarks show our technique improves performance on average by 1.7X with energy savings of 30%. We further study the performance sensitivity of DWS to various D-cache sizes and associativities. The performance benefit of DWS is also demonstrated with various SIMD widths and multi-threading depths. While DWS may increase the number of scheduling entities, we propose ways to limit this side effect. The effectiveness of such technique is also analyzed quantitatively." @default.
• W2321262893 created "2016-06-24" @default.
• W2321262893 creator A5018781096 @default.
• W2321262893 creator A5061578459 @default.
• W2321262893 creator A5074818897 @default.
• W2321262893 date "2010-01-01" @default.
• W2321262893 modified "2023-09-27" @default.
• W2321262893 title "Dynamic Warp Subdivision for Integrated Branch and Memory Divergence Tolerance: Extended Tradeoff Analysis" @default.
• W2321262893 cites W1525740151 @default.
• W2321262893 cites W1981319267 @default.
• W2321262893 cites W1985722570 @default.
• W2321262893 cites W2012252449 @default.
• W2321262893 cites W2037743346 @default.
• W2321262893 cites W2044206819 @default.
• W2321262893 cites W2097909406 @default.
• W2321262893 cites W2102727118 @default.
• W2321262893 cites W2105012172 @default.
• W2321262893 cites W2106625514 @default.
• W2321262893 cites W2107978915 @default.
• W2321262893 cites W2108977887 @default.
• W2321262893 cites W2120692212 @default.
• W2321262893 cites W2120964511 @default.
• W2321262893 cites W2122646225 @default.
• W2321262893 cites W2128022558 @default.
• W2321262893 cites W2132587889 @default.
• W2321262893 cites W2144481293 @default.
• W2321262893 cites W2145021036 @default.
• W2321262893 cites W2145866640 @default.
• W2321262893 cites W2146246439 @default.
• W2321262893 cites W2148041475 @default.
• W2321262893 cites W2151982028 @default.
• W2321262893 cites W2156831150 @default.
• W2321262893 cites W2164333604 @default.
• W2321262893 cites W2168452045 @default.
• W2321262893 cites W2169150396 @default.
• W2321262893 cites W2169880332 @default.
• W2321262893 cites W2170879098 @default.
• W2321262893 cites W2535359146 @default.
• W2321262893 cites W1567324076 @default.
• W2321262893 hasPublicationYear "2010" @default.
• W2321262893 type Work @default.
• W2321262893 sameAs 2321262893 @default.
• W2321262893 citedByCount "0" @default.
• W2321262893 crossrefType "journal-article" @default.
• W2321262893 hasAuthorship W2321262893A5018781096 @default.
• W2321262893 hasAuthorship W2321262893A5061578459 @default.
• W2321262893 hasAuthorship W2321262893A5074818897 @default.
• W2321262893 hasConcept C111919701 @default.
• W2321262893 hasConcept C113166858 @default.
• W2321262893 hasConcept C115537543 @default.
• W2321262893 hasConcept C117280010 @default.
• W2321262893 hasConcept C138101251 @default.
• W2321262893 hasConcept C162324750 @default.
• W2321262893 hasConcept C16320812 @default.
• W2321262893 hasConcept C173608175 @default.
• W2321262893 hasConcept C189783530 @default.
• W2321262893 hasConcept C202491316 @default.
• W2321262893 hasConcept C206729178 @default.
• W2321262893 hasConcept C21547014 @default.
• W2321262893 hasConcept C28034677 @default.
• W2321262893 hasConcept C38556500 @default.
• W2321262893 hasConcept C41008148 @default.
• W2321262893 hasConcept C76155785 @default.
• W2321262893 hasConcept C82876162 @default.
• W2321262893 hasConceptScore W2321262893C111919701 @default.
• W2321262893 hasConceptScore W2321262893C113166858 @default.
• W2321262893 hasConceptScore W2321262893C115537543 @default.
• W2321262893 hasConceptScore W2321262893C117280010 @default.
• W2321262893 hasConceptScore W2321262893C138101251 @default.
• W2321262893 hasConceptScore W2321262893C162324750 @default.
• W2321262893 hasConceptScore W2321262893C16320812 @default.
• W2321262893 hasConceptScore W2321262893C173608175 @default.
• W2321262893 hasConceptScore W2321262893C189783530 @default.
• W2321262893 hasConceptScore W2321262893C202491316 @default.
• W2321262893 hasConceptScore W2321262893C206729178 @default.
• W2321262893 hasConceptScore W2321262893C21547014 @default.
• W2321262893 hasConceptScore W2321262893C28034677 @default.
• W2321262893 hasConceptScore W2321262893C38556500 @default.
• W2321262893 hasConceptScore W2321262893C41008148 @default.
• W2321262893 hasConceptScore W2321262893C76155785 @default.
• W2321262893 hasConceptScore W2321262893C82876162 @default.
• W2321262893 hasLocation W23212628931 @default.
• W2321262893 hasOpenAccess W2321262893 @default.
• W2321262893 hasPrimaryLocation W23212628931 @default.
• W2321262893 hasRelatedWork W1987414320 @default.
• W2321262893 hasRelatedWork W1988625252 @default.
• W2321262893 hasRelatedWork W2012303498 @default.
• W2321262893 hasRelatedWork W2139768739 @default.
• W2321262893 hasRelatedWork W2146849645 @default.
• W2321262893 hasRelatedWork W2151147384 @default.
• W2321262893 hasRelatedWork W2156831150 @default.
• W2321262893 hasRelatedWork W2160106616 @default.
• W2321262893 hasRelatedWork W2161101294 @default.
• W2321262893 hasRelatedWork W2304172997 @default.
• W2321262893 hasRelatedWork W2536535850 @default.
• W2321262893 hasRelatedWork W2610229489 @default.
• W2321262893 hasRelatedWork W2803678556 @default.
• W2321262893 hasRelatedWork W2936814528 @default.
• W2321262893 hasRelatedWork W2953836263 @default.
• W2321262893 hasRelatedWork W2996477311 @default.

• next