Tootfinder

Opt-in global Mastodon full text search. Join the index!

No exact results. Similar results found.
@Techmeme@techhub.social
2026-06-29 01:50:38

Jefferies: memory prices are expected to rise 40-50% in Q3 2026 and 30-40% in Q4, as Chinese memory makers are unlikely to provide meaningful near-term relief (Hassan Mujtaba/Wccftech)
wccftech.com/jefferies-warns-m

@raiders@darktundra.xyz
2026-09-07 22:45:45

Watch: Every run in Raiders RB Mike Washington's NFL leading preseason raiderswire.usatoday.com/story

@memeorandum@universeodon.com
2026-06-25 17:01:17

Eager for a break, farmers don't expect much relief after the Iran peace deal (Kirk Siegler/NPR)
npr.org/2026/06/25/nx-s1-58676
memeorandum.com/260625/p70#a26

@arXiv_csAR_bot@mastoxiv.page
2026-08-13 07:39:50

NITRO: High-Performance 3D NAND Flash-Based In-Storage Computing with Enhanced Activation Dataflow
Sanghun Shin, Sangyeon Kim, Gisan Ji, Sungju Ryu
arxiv.org/abs/2608.11920 arxiv.org/pdf/2608.11920 arxiv.org/html/2608.11920
arXiv:2608.11920v1 Announce Type: new
Abstract: In-storage computing (ISC) is considered a next-generation memory architecture for its ability to relieve the data bottleneck between the host and the memory. While the required resources of large language models (LLMs) have increased significantly in recent years, the memory density has not scaled accordingly. Recently, several works have studied NAND flash-based processing-in-memory (NAND-PIM) schemes to exploit the high density of the memory. However, they do not address the dataflow/buffer for the intermediate values, so a simple method is to deal with the values in the slow flash memory array. To overcome such a limitation, we propose a high-performance NAND flash-based ISC architecture with enhanced activation buffering. Instead of using the very slow flash memory array for the intermediate values, our architecture buffers the values in a fast DRAM subsystem. This approach effectively handles the high-latency penalties when activations are programmed into slower TLC NAND flash. We also introduce a distributed dataflow approach for the NAND-PIM array. This approach maximizes computational parallelism by employing efficient intra-plane data mapping. The results show that our proposed architecture achieves significant performance improvements, reducing the inference latency by up to 85% compared to the baseline.
toXiv_bot_toot

@primonatura@mstdn.social
2026-08-07 17:00:58

"In coastal Japan, one project is planting forests to stand against tsunamis"
#Japan #Tsunamis #Trees #Environment

@Techmeme@techhub.social
2026-06-25 12:47:32

Apple raises Mac prices by 15% to 20% and iPad prices by 15% to 25%, saying it has "never seen a component price increase this much, this quickly" (Rolfe Winkler/Wall Street Journal)
wsj.com/tech/apple-r…

Trump’s fragile agreement with Iran has introduced a new variable into this year’s midterm elections.
Democratic candidates have assailed the agreement,
arguing that the president accepted unfavorable terms to try to end an unnecessary war that hurt the economy.
Republicans have been more divided.
Some, eager to turn the page on the unpopular war, are hailing falling gas prices and praising Trump for weakening Iran’s military capabilities,
while others are expr…

@raiders@darktundra.xyz
2026-08-28 03:28:32

Fernando Mendoza shaky in preseason finale: Instant analysis, grade usatoday.com/story/sports/nfl/

@raiders@darktundra.xyz
2026-08-21 16:31:44

Highlights: Watch the best moments from the Raiders' 22-20 preseason win over the Texans raiders.com/news/highlights-wa

@raiders@darktundra.xyz
2026-08-21 21:28:32

Highlights: Watch the best moments from the Raiders' 22-20 preseason win over the Texans raiders.com/news/highlights-wa