[B! GPU] pochi-mkã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

pochi-mk id:pochi-mk

GPUã«é–¢ã™ã‚‹pochi-mkã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (12)

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

${{author_name}}$
{{author_name}}{{created}}
{{ #comment }}{{ comment }}{{ /comment }}
- {{ label }}

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

Google Cloud TPUã‚’ä½¿ã£ãŸè¨ˆé‡å¦ç¿’ã®é«˜é€ŸåŒ–äº‹ä¾‹ã®ç´¹ä»‹ - ZOZO TECH BLOG
ZOZOç ”ç©¶æ‰€ã®å¾Œè—¤ã§ã™ã€‚æœ¬è¨˜äº‹ã§ã¯Google Cloud TPUã‚’ä½¿ã£ãŸè¨ˆé‡å¦ç¿’ã®é«˜é€ŸåŒ–ã®äº‹ä¾‹ã‚’ç´¹ä»‹ã—ã¾ã™ã€‚ ã¯ã˜ã‚ã« æ·±å±¤å¦ç¿’ã‚’ç”¨ã„ãŸãƒ—ãƒãƒ€ã‚¯ãƒˆã‚’é–‹ç™ºãƒ»é‹ç”¨ã™ã‚‹ä¸Šã§ã€ãƒ¢ãƒ‡ãƒ«ã®å¦ç¿’ã«ã‹ã‹ã‚‹è†¨å¤§ãªæ™‚é–“ã¯ãƒœãƒˆãƒ«ãƒãƒƒã‚¯ã®1ã¤ã§ã™ã€‚ ãƒ•ã‚¡ãƒƒã‚·ãƒ§ãƒ³ã«ãŠã‘ã‚‹æ·±å±¤å¦ç¿’ã‚’ç”¨ã„ãŸç”»åƒèªè˜æŠ€è¡“ã«ã‚‚åŒã˜ã“ã¨ãŒè¨€ãˆã¾ã™ã€‚ ä»Šå›žã¯ãƒ•ã‚¡ãƒƒã‚·ãƒ§ãƒ³ã®åˆ†é‡Žã«ãŠã„ã¦å®šç•ªã®ã‚¿ã‚¹ã‚¯ã§ã‚ã‚‹Street2shopã®èª²é¡Œè¨å®šã«å¯¾ã—ã€Google Cloud TPUã‚’ç”¨ã„ã¦è¨ˆé‡å¦ç¿’ã®é«˜é€ŸåŒ–ã‚’è©¦ã¿ã¾ã™ã€‚ Street2shopã¯ã€ã‚¹ãƒŠãƒƒãƒ—ç”»åƒã‹ã‚‰å•†å“éƒ¨åˆ†ã‚’åˆ‡ã‚Šå‡ºã™ç‰©ä½“æ¤œå‡ºã®ãƒ‘ãƒ¼ãƒˆã¨ã€åˆ‡ã‚Šå‡ºã—ãŸç”»åƒã¨é¡žä¼¼ã—ãŸå•†å“ã‚’æ¤œç´¢ã™ã‚‹ã‚¯ãƒã‚¹ãƒ‰ãƒ¡ã‚¤ãƒ³ç”»åƒæ¤œç´¢ã®ãƒ‘ãƒ¼ãƒˆã«åˆ†ã‘ã‚‰ã‚Œã¾ã™ã€‚ ä»Šå›žã®å–ã‚Šçµ„ã¿ã§ã¯ã€å¾Œè€…ã®ãƒ‘ãƒ¼ãƒˆã§åˆ©ç”¨ã™ã‚‹ç”»åƒé–“ã®è·é›¢ã‚’æ¸¬ã‚‹ãŸã‚ã®ãƒ¢ãƒ‡ãƒ«ã®å¦ç¿’ã®é«˜é€ŸåŒ–ã‚’è¡Œã„ã¾ã™ã€‚ ç›®æ¬¡ ã¯ã˜ã‚ã« ç›®æ¬¡ Google Cloud TPUã¨ã¯ ã‚¿ã‚¹ã‚¯ S
pochi-mk 2019/02/12
ã€Œä»Šå›žã®è¨ˆé‡å¦ç¿’ã®ãŸã‚ã«ã€å•†å“ç”»åƒã¨ç€ç”¨æ™‚ã®ç”»åƒã®ãƒšã‚¢ã‚’120ä¸‡çµ„ã»ã©ç”¨æ„ã—ã¾ã—ãŸã€‚ã€ã¤ã‚ˆã„ã€‚

TensorFlow

GPU

cloud

google

ç”»åƒ
ãƒªãƒ³ã‚¯
TensorFlow LiteãŒGPUã‚’ã‚µãƒãƒ¼ãƒˆã—ãŸã‚‰ã—ã„ã®ã§iOSã§è©¦ã—ã¦ã¿ãŸ - Qiita
TensorFlow LiteãŒAndroidã‚„iOSãƒ‡ãƒã‚¤ã‚¹ã®GPUã‚’ã‚µãƒãƒ¼ãƒˆã—ãŸã‚ˆã†ã§ã™1ã€‚ TensorFlow Lite Now Faster with Mobile GPUs (Developer Preview) ãƒãƒƒã‚¯ã‚¨ãƒ³ãƒ‰ã¯Androidã§ã¯OpenGL ES 3.1ã€iOSã§ã¯Metalã€‚ This new backend leverages: OpenGL ES 3.1 Compute Shaders on Android devices Metal Compute Shaders on iOS devices iPhone 7ã§ã¯æœ€å¤§6å€ã®é«˜é€ŸåŒ–ãŒè¦‹ã‚‰ã‚ŒãŸã‚ˆã†ã§ã™ã€‚ By leveraging the new GPU backend in the future, inference can be sped up from ~4x on Pixel 3 and Sams
pochi-mk 2019/01/20
iOS

GPU

TensorFlow
ãƒªãƒ³ã‚¯
é€²æ—ï¼‰SSD-to-GPU ãƒ€ã‚¤ãƒ¬ã‚¯ãƒˆSQLå®Ÿè¡Œæ©Ÿèƒ½ - KaiGaiã®ä¿ºãƒ¡ãƒ¢
ã“ã“æš«ããƒ–ãƒã‚°ã§ã¾ã¨ã‚ã¦ã„ãªã‹ã£ãŸã€SSD-to-GPUãƒ€ã‚¤ãƒ¬ã‚¯ãƒˆSQLå®Ÿè¡Œæ©Ÿèƒ½ã®é€²æ—ã«ã¤ã„ã¦ã€‚ ã“ã®æ©Ÿèƒ½ã‚’ã‹ã„ã¤ã¾ã‚“ã§è¨€ã†ã¨ã€NVMe-SSDã«æ ¼ç´ã•ã‚Œã¦ã„ã‚‹PostgreSQLã®ãƒ‡ãƒ¼ã‚¿ãƒ–ãƒãƒƒã‚¯ã‚’GPU RAMã«ç›´æŽ¥è»¢é€ã—ã€ãã“ã§SQLã®WHEREå¥/JOIN/GROUP BYã‚’å®Ÿè¡Œã™ã‚‹ã“ã¨ã§è¦‹ã‹ã‘ä¸Šã®I/Oé‡ã‚’å‰Šæ¸›ã™ã‚‹ã¨ã„ã†ä»£ç‰©ã§ã‚ã‚‹ã€‚ NVIDIAã®Tesla/Quadro GPUãŒå¯¾å¿œã™ã‚‹GPUDirect RDMAæ©Ÿèƒ½ã‚’ä½¿ã„ã€SSD<=>GPUé–“ã®ãƒ‡ãƒ¼ã‚¿è»¢é€ã‚’ä»²ä»‹ã™ã‚‹Linux kernel moduleã‚’ä½¿ãˆã°ã€CPU/RAMã«ãƒ‡ãƒ¼ã‚¿ã‚’ãƒãƒ¼ãƒ‰ã™ã‚‹å‰ã«GPUä¸Šã§ã®å‡¦ç†ã‚’è¡Œã†ã“ã¨ãŒã§ãã‚‹ã€‚ ã—ã°ã‚‰ãå‰ã‹ã‚‰Scanç³»ã®å‡¦ç†ã«ã¯å¯¾å¿œã—ã¦ã„ãŸãŒã€JOIN/GROUP BYã¸ã®å¯¾å¿œã‚’åŠ ãˆã€ã•ã‚‰ã«PostgreSQL v9.6ã®CPUä¸¦åˆ—ã«ã‚‚è¿½å¾“ã—ãŸã¨ã„ã†ã“ã¨ã§ã€ç°¡å˜ãªãƒ™ãƒ³ãƒãƒžãƒ¼ã‚¯ãªã‚‰å–ã‚Œã‚‹
pochi-mk 2017/04/24
ssd

GPU

PostgreSQL

ãƒ‡ãƒ¼ã‚¿
ãƒªãƒ³ã‚¯
Keras.js â€“ æ©Ÿæ¢°å¦ç¿’ã‚’ãƒ–ãƒ©ã‚¦ã‚¶ä¸Šã§å®Ÿè¡Œã§ãã‚‹JSãƒ©ã‚¤ãƒ–ãƒ©ãƒªã€GPUãƒ‘ãƒ¯ãƒ¼ã‚’ä½¿ç”¨ ã€@maskinã€‘ | TechWaveï¼ˆãƒ†ãƒƒã‚¯ã‚¦ã‚§ãƒ¼ãƒ–ï¼‰
1990å¹´ä»£åˆé ã‹ã‚‰è¨˜è€…ã¨ã—ã¦ã¾ãŸèµ·æ¥å®¶ã¨ã—ã¦ITã‚¹ã‚¿ãƒ¼ãƒˆã‚¢ãƒƒãƒ—æ¥ç•Œã®ãƒãƒ¼ãƒ‰ã‚¦ã‚§ã‚¢ã‹ã‚‰ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã®äº‹æ¥å‰µå‡ºã«é–¢ã‚ã‚‹ã€‚ã‚·ãƒªã‚³ãƒ³ãƒãƒ¬ãƒ¼ã‚„EUç‰ã§ã®ã‚¹ã‚¿ãƒ¼ãƒˆã‚¢ãƒƒãƒ—ã‚’çµŒé¨“ã€‚æ—¥æœ¬ã§ã¯ãƒãƒƒãƒˆã‚¨ã‚¤ã‚¸ç‰ã«æ‰€å±žã€å¤§æ‰‹ä¼æ¥ã®æ–°è¦äº‹æ¥å‰µå‡ºã«å”åŠ›ã€‚ãƒ–ãƒã‚°ã‚„SNSã€LINEãªã©ã®èª•ç”Ÿã‹ã‚‰æ™®åŠæˆé•·ã¾ã§ã‚’æœ€å‰ç·šã§è¦‹ã¦ããŸç”Ÿãå—å¼•ã¨ã—ã¦æ³¨ç›®ã•ã‚Œã‚‹ã€‚é€šä¿¡ã‚ãƒ£ãƒªã‚¢ã®ãƒ‹ãƒ¥ãƒ¼ã‚¹ãƒãƒ¼ã‚¿ãƒ«ã®å‰µæ¥ãƒ‡ã‚¹ã‚¯ã¨ã—ã¦æ•°å„„PVäº‹æ¥ã«ã€‚ä¸–ç•Œæœ€å¤§ITç³»ãƒ¡ãƒ‡ã‚£ã‚¢ï¼ˆã‚¹ãƒšã‚¤ãƒ³ï¼‰ã®å…ƒæ—¥æœ¬ç·¨é›†é•·ã€World Innovation Lab(WiL)ãªã©ã‚’çµŒã¦ã€ç¾åœ¨ã€ã‚¹ã‚¿ãƒ¼ãƒˆã‚¢ãƒƒãƒ—æ”¯æ´å´ã®å–ã‚Šçµ„ã¿ã«æ³¨åŠ›ä¸ã€‚ Pythonã§æ›¸ã‹ã‚ŒãŸãƒ‡ã‚£ãƒ¼ãƒ—ãƒ©ãƒ¼ãƒ‹ãƒ³ã‚°ç”¨ãƒ‹ãƒ¥ãƒ¼ãƒ©ãƒ«ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ãƒ©ã‚¤ãƒ–ãƒ©ãƒªã€ŒKerasã€ã‚’ã€ã‚¦ã‚§ãƒ–ãƒ–ãƒ©ã‚¦ã‚¶ä¸Šã§å®Ÿè¡Œã™ã‚‹JavaScriptãƒ©ã‚¤ãƒ–ãƒ©ãƒªã€ŒKeras.jsã€ãŒç™»å ´ã—ãŸã€‚ ãã‚‚ãã‚‚ã€ŒKerasã€ã¯ã€GoogleãŒæä¾›ã™ã‚‹ã‚ªãƒ¼ãƒ—ãƒ³ã‚½ãƒ¼ã‚¹ã®æ©Ÿæ¢°å¦ç¿’ãƒ»
pochi-mk 2016/10/14
ã‚ã¨ã§èªã‚€

GPU

network

ãƒ©ã‚¤ãƒ–ãƒ©ãƒª

æ©Ÿæ¢°å¦ç¿’

ãƒ‡ã‚£ãƒ¼ãƒ—ãƒ©ãƒ¼ãƒ‹ãƒ³ã‚°
ãƒªãƒ³ã‚¯
AlphaGoã®ã—ãã¿
SSII2021 [TS2] æ·±å±¤å¼·åŒ–å¦ç¿’ ã€œ å¼·åŒ–å¦ç¿’ã®åŸºç¤Žã‹ã‚‰å¿œç”¨ã¾ã§ ã€œ 6/10 (æœ¨) 9:30ï½ž10:40 è¬›å¸«ï¼šå¹³å· ç¿¼ æ°ï¼ˆä¸éƒ¨å¤§å¦ï¼‰ æ¦‚è¦ï¼š æ·±å±¤å¼·åŒ–å¦ç¿’ã¯Deep Q-Network (DQN) ã®ç™»å ´ä»¥é™ã€æ§˜ã€…ãªã‚¢ãƒ—ãƒãƒ¼ãƒãŒææ¡ˆã•ã‚Œã¦ãŠã‚Šã€Alpha Goã«ã‚ˆã‚‹å›²ç¢ã®æ”»ç•¥ã‚„ãƒãƒœãƒƒãƒˆã®è‡ªå¾‹åˆ¶å¾¡ãªã©ã®æ§˜ã€…ãªå¿œç”¨ãŒãªã•ã‚Œã¦ã„ã¾ã™ã€‚æœ¬ãƒãƒ¥ãƒ¼ãƒˆãƒªã‚¢ãƒ«ã§ã¯ã€å¾“æ¥ã®å¼·åŒ–å¦ç¿’ã®åŸºæœ¬çš„ãªè€ƒãˆæ–¹ã«è§¦ã‚Œã€æ·±å±¤å¦ç¿’ã‚’çµ„ã¿åˆã‚ã›ãŸæ·±å±¤å¼·åŒ–å¦ç¿’ã«ã¤ã„ã¦ã®ç´¹ä»‹ã‚’è¡Œã„ã¾ã™ã€‚ã¾ãŸã€æ™‚é–“ã®è¨±ã™é™ã‚Šã€æœ€æ–°ã®æ·±å±¤å¦ç¿’æ‰‹æ³•ã‚„Alpha Goã®ä»•çµ„ã¿ã€æ·±å±¤å¼·åŒ–å¦ç¿’ã®æ´»ç”¨ä¾‹ã‚’ç´¹ä»‹ã—ã¾ã™ã€‚
pochi-mk 2016/04/27
GPU

Deep Learning

SlideShare
ãƒªãƒ³ã‚¯
ã“ã‚Œã‹ã‚‰ã®ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ãƒ†ã‚£ãƒ³ã‚°ã®å¤‰åŒ–ã¨Java-JJUG CCC 2015 Fall
11. ãƒ‡ãƒ¼ã‚¿ãƒ»ã‚»ãƒ³ãƒˆãƒªãƒƒã‚¯ãƒ»ã‚·ã‚¹ãƒ†ãƒ â— ãƒ‡ãƒ¼ã‚¿ã®ç§»å‹•ã«é›»åŠ›ã‚„æ™‚é–“ãŒé£Ÿã‚ã‚Œã¦ã„ã‚‹ â€“ ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸â†’ãƒ¡ã‚¤ãƒ³ãƒ¡ãƒ¢ãƒªâ†’ã‚ãƒ£ãƒƒã‚·ãƒ¥ãƒ¡ãƒ¢ãƒª â— ãƒ‡ãƒ¼ã‚¿ã®ç§»å‹•ã‚’æ¸›ã‚‰ã™å¿…è¦ãŒã‚ã‚‹ â— ãƒ‡ãƒ¼ã‚¿ã®è¿‘ãã§å‡¦ç†ã‚’è¡Œã† â— å‡¦ç†ã‚’è¡Œã†ã®ã¯CPUã ã‘ã§ã¯ãªããªã‚‹ â— ã„ã‚ã„ã‚ãªã¨ã“ã‚ã§å‹•ãã‚³ãƒ¼ãƒ‰ãŒå¿…è¦ã«ãªã‚‹
pochi-mk 2015/12/07
JJUG

Java

cpu

GPU

kis
ãƒªãƒ³ã‚¯
scryptãŒGPUã«ç ´ã‚‰ã‚Œã‚‹æ™‚ | ã³ã‚Šã‚ã‚‹ã®ç ”ç©¶ãƒŽãƒ¼ãƒˆ
ä¸€èˆ¬çš„ã«ã‚ˆãçŸ¥ã‚‰ã‚Œã¦ã„ã‚‹ SHA-256 ã‚„ MD5 ãªã©ã®ãƒãƒƒã‚·ãƒ¥é–¢æ•°ã¯éžå¸¸ã«å˜ç´”ãªè¨è¨ˆã¨ãªã£ã¦ãŠã‚Šã€éžåŠ›ãªãƒ‘ã‚½ã‚³ãƒ³ã‚„çµ„ã¿è¾¼ã¿æ©Ÿå™¨ã€ã‚¹ãƒžãƒ•ã‚©ãªã©ã§ã‚‚é«˜é€Ÿã«è¨ˆç®—ã§ãã¾ã™ã€‚ ã—ã‹ã—ãªãŒã‚‰ãã®ä¸€æ–¹ã§ã€ãƒãƒƒã‚·ãƒ¥é–¢æ•°ã‚’æ‰‹å½“ãŸã‚Šæ¬¡ç¬¬ã«è¨ˆç®—ã—ã€ã‚‚ã¨ã®å…¥åŠ›å€¤ã‚’å¾©å…ƒã™ã‚‹ã„ã‚ã‚†ã‚‹ã€Œãƒ–ãƒ«ãƒ¼ãƒˆãƒ•ã‚©ãƒ¼ã‚¹æ”»æ’ƒã€ãŒå®¹æ˜“ã§ã‚ã‚‹ã¨ã„ã†ãƒ‡ãƒ¡ãƒªãƒƒãƒˆãŒã‚ã‚Šã¾ã™ã€‚ ç‰¹ã«ã“ã®ã‚ˆã†ãª SHA-256 ã‚„ MD5 ã¨ã„ã£ãŸãƒãƒƒã‚·ãƒ¥é–¢æ•°ã¯ã€GPU ã‚’ç”¨ã„ã‚‹ã‹ã€ã‚‚ã—ãã¯å°‚ç”¨ã®ãƒãƒ¼ãƒ‰ã‚¦ã‚§ã‚¢ (FPGA ã‚‚ã—ãã¯ ASIC) ã‚’è£½ä½œã™ã‚‹ã“ã¨ã§éžå¸¸ã«é«˜ã„åŠ¹çŽ‡ã§è¨ˆç®—ï¼ˆæ”»æ’ƒï¼‰ãŒã§ãã¦ã—ã¾ã†ã“ã¨ãŒçŸ¥ã‚‰ã‚Œã¦ã„ã¾ã™ã€‚ ãã®ãŸã‚ã€GPU ãªã„ã—å°‚ç”¨ãƒãƒ¼ãƒ‰ã‚¦ã‚§ã‚¢ã‚’ç”¨ã„ãŸã¨ã—ã¦ã‚‚ã€æ”»æ’ƒåŠ¹çŽ‡ã®æ”¹å–„ãŒé›£ã—ããªã‚‹ã‚ˆã†ãªæ–°ãŸãªãƒãƒƒã‚·ãƒ¥é–¢æ•°ãŒã„ãã¤ã‹ææ¡ˆã•ã‚Œã¦ã„ã¾ã™ã€‚ ãã®ä¸ã§æ¯”è¼ƒçš„å¤ã (2012å¹´ã”ã‚) ã«é–‹ç™ºã•ã‚Œã€ä»–ã®ãƒãƒƒã‚·ãƒ¥é–¢æ•°ã«ã‚‚å½±éŸ¿ã‚’ä¸Žãˆã¦ã„ã‚‹ã€Œscrypt
pochi-mk 2015/10/12
GPU

ã‚ã¨ã§èªã‚€

ã‚»ã‚ãƒ¥ãƒªãƒ†ã‚£

scrypt

security
ãƒªãƒ³ã‚¯
COOL Chips XVIII - Wisconsin-Madisonå¤§å¦ã®Open GPUã€ŒMiaowã€
COOL Chips XVIIIã«ãŠã„ã¦Wisonsin-Madionå¤§å¦ã®ã‚°ãƒ«ãƒ¼ãƒ—ãŒã€ŒMIAOWã€ã¨å‘¼ã¶ã‚ªãƒ¼ãƒ—ãƒ³ãªGPUã‚’ç™ºè¡¨ã—ãŸã€‚RTLãªã©ãŒå…¬é–‹ã•ã‚Œã¦ã„ã‚‹CPUã¯ã€ãƒ¡ãƒ¼ã‚«ãƒ¼ãŒå…¬é–‹ã—ã¦ã„ã‚‹OpenSPARCã‚„LEON SPARCã‚„å¤§å¦ãªã©ãŒé–‹ç™ºã—ãŸFabScalarãªã©å„ç¨®ã®ã‚‚ã®ãŒã‚ã‚Šã€ã“ã‚Œã‚‰ã‚’ãƒ™ãƒ¼ã‚¹ã«ãƒžã‚¤ã‚¯ãƒã‚¢ãƒ¼ã‚ãƒ†ã‚¯ãƒãƒ£ã‚’å¤‰æ›´ã—ã¦ã€ãã®åŠ¹æžœã‚’æ¤œè¨¼ã™ã‚‹ã¨ã„ã†ç ”ç©¶ãŒã§ãã‚‹ã®ã§ã‚ã‚‹ãŒã€GPUã§ã¯ã“ã®ã‚ˆã†ãªã‚ªãƒ¼ãƒ—ãƒ³ãªã‚‚ã®ãŒç„¡ãã€ç ”ç©¶ç’°å¢ƒãŒãªã„ã€‚MIAOWã¯ã€RTLã‚’å…¬é–‹ã™ã‚‹ã‚ªãƒ¼ãƒ—ãƒ³ãªGPUã‚’æä¾›ã™ã‚‹ã“ã¨ã‚’ç›®çš„ã¨ã—ã¦ãŠã‚Šã€ã“ã®RTLã‚’è‡ªç”±ã«å¤‰æ›´ã—ã¦GPUã‚¢ãƒ¼ã‚ãƒ†ã‚¯ãƒãƒ£ã®ç ”ç©¶ã‚’è¡Œã†ã“ã¨ãŒã§ãã‚‹ã€‚ MIAOW(æ—¥æœ¬èªžã§ã¯ãƒŸãƒ£ãƒ¼ã‚ª)ã¯ã€ŒMany-core Integrated Accelerator Of Wisconsinã€ã®é æ–‡å—ã‚’ã¨ã£ãŸåå‰ã§ã‚ã‚‹ã€‚ã‚°ãƒ©ãƒ•ã‚£ãƒƒã‚¯ã‚¹å°‚ç”¨æ©Ÿèƒ½ã¯å«ã‚“ã§ã„ãªã„ãŒAMD
pochi-mk 2015/05/15
GPU

ç ”ç©¶

ãƒãƒ¼ãƒ‰ã‚¦ã‚§ã‚¢

ã‚¢ãƒ¼ã‚ãƒ†ã‚¯ãƒãƒ£

FPGA

ã‚ªãƒ¼ãƒ—ãƒ³ã‚½ãƒ¼ã‚¹
ãƒªãƒ³ã‚¯
OpenCLã‚„ã‚‹å‰ã«SIMDä½¿ã„åˆ‡ã‚Œã£ã¦ã„ã†å¹»æƒ³ - aokomoriuta's blog
ãŠè©«ã³ ã‚¢ãƒ©ã‚¤ãƒ³å¿˜ã‚Œã¦ã¾ã—ãŸã”ã‚ã‚“ãªã•ã„ã€ã§ã‚‚æ™‚é–“ã‚ã‚“ã¾ã‚Šå¤‰ã‚ã£ã¦ãªã„ã‹ã‚‰è¨±ã—ã¦ãƒ»ãƒ»ãƒ» https://t.co/JNtq2U2kMqâ€” é’åå®ˆæŒ (@aokomoriuta) April 29, 2015 ã§ã¯æœ¬ç·¨ã©ã†ãžâ†“ æœ¬ç·¨ è‹¥å¹²è©±é¡Œã«ä¹—ã‚Šé…ã‚ŒãŸæ„Ÿã‚ã‚Šã¾ã™ãŒã€‚ d.hatena.ne.jp ã‘ã©ã€SSEã‚‚çŸ¥ã‚‰ããƒ¼ã€SIMDã‚‚çŸ¥ã‚‰ããƒ¼ã€ãªã‚“ã‹ä¿ºãŒæ›¸ã„ãŸã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ é…ã„ã‘ã©ã¨ã‚Šã‚ãˆãšOpenCLã¨ã‹ã§é«˜é€ŸåŒ–ã—ã‚ˆã£ã‹ãªãƒ¼ã¨ã‹ãã€ç”˜ã£ãŸã‚Œã¦ã‚“ã˜ã‚ƒããˆã‚ˆã€‚CPUã‚’ã‚‚ã£ã¨ä½¿ã„ãã‚Œã‚ˆã€‚ãŠå‰ã®ã‚¢ã‚ªã„ã‚³ãƒ¼ãƒ‰ã®ã›ã„ã§CPUãŒæ³£ã„ã¦ã‚‹ã‚ˆã€‚ã£ã¦ã„ã†è©±ã§ã™ã‚ˆã€‚ GPGPUãªã‚“ã¦ã®ã¯SIMDã‚’ä½¿ã„åˆ‡ã£ãŸå¾Œã®è©±ã§ã—ã‚‡ã€‚ GPGPUã™ã‚‹ã®ã«GPUã®ãƒ‘ãƒ¯ãƒ¼ã¨ãƒ¡ãƒ¢ãƒªãŒè¶³ã‚Šã¾ã›ã‚“ã¨ã‹è¨€ã†å‰ã«ã¾ãšSIMDã‹ã‚‰ã ã‚ã€‚ ã¨ã‹è¨€ã‚ã‚Œã¦ãŸã®ã§ã€æ¤œè¨¼ã™ã‚‹ã“ã¨ã«ã—ãŸï¼ˆã‚„ã£ã¤ã‘ï¼‰ã€‚ ç’°å¢ƒ OS: Windows 7 Profession
pochi-mk 2015/04/29
GPU

c++

GPGPU
ãƒªãƒ³ã‚¯
ã€å¾Œè—¤å¼˜èŒ‚ã®Weeklyæµ·å¤–ãƒ‹ãƒ¥ãƒ¼ã‚¹ã€‘ NVIDIAãŒç©¶æ¥µã®ãƒ¢ãƒã‚¤ãƒ«SoCã€ŒTegra K1ã€ã‚’ç™ºè¡¨
pochi-mk 2014/01/07
ã€Œãƒ¢ãƒã‚¤ãƒ«ã‹ã‚‰ã‚¹ãƒ¼ãƒ‘ãƒ¼ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ã‚¿ã¾ã§ã®ã‚¢ãƒ¼ã‚ãƒ†ã‚¯ãƒãƒ£çµ±åˆã€ã£ã¦ã€ã‹ã¤ã¦ Sun ãŒ SPARC ã§å®Ÿç¾ã—ã‚ˆã†ã¨ã—ã¦ã„ãŸå¤¢ã ã£ãŸã‚ˆã†ãª...

NVIDIA

ãƒ¢ãƒã‚¤ãƒ«

GPU

cpu

ã‚¢ãƒ¼ã‚ãƒ†ã‚¯ãƒãƒ£

hardware
ãƒªãƒ³ã‚¯
å††å‘¨çŽ‡ã®å°æ•°ç‚¹ä»¥ä¸‹8000å…†æ¡ã‚ã‚’GeForceã§æ±‚ã‚ã‚‹æ–¹æ³•ã€œGTC 2013ã®ã€ŒÏ€ã€ã‚»ãƒƒã‚·ãƒ§ãƒ³ãƒ¬ãƒãƒ¼ãƒˆ
å††å‘¨çŽ‡ã®å°æ•°ç‚¹ä»¥ä¸‹8000å…†æ¡ã‚ã‚’GeForceã§æ±‚ã‚ã‚‹æ–¹æ³•ã€œGTC 2013ã®ã€ŒÏ€ã€ã‚»ãƒƒã‚·ãƒ§ãƒ³ãƒ¬ãƒãƒ¼ãƒˆ ãƒ©ã‚¤ã‚¿ãƒ¼ï¼šè¥¿å·å–„å¸ Ed Karrelsæ° ã‚»ãƒƒã‚·ãƒ§ãƒ³ã‚¿ã‚¤ãƒˆãƒ« æ•°å¦è€…ã®é–“ã§ã€Œå††å‘¨çŽ‡ã®æ—¥ã€ã¨å‘¼ã°ã‚Œã‚‹3æœˆ14æ—¥ã€‚2013å¹´ã®ã“ã®æ—¥ã«ï¼Œç±³å›½ã‚µãƒ³ã‚¿ã‚¯ãƒ©ãƒ©å¤§å¦ï¼ŒEd Karrelsæ°ã‚‰ã®ç ”ç©¶ã‚°ãƒ«ãƒ¼ãƒ—ã¯ï¼Œå††å‘¨çŽ‡è¨ˆç®—ã«é–¢ã™ã‚‹æ–°è¨˜éŒ²ã€Œå°æ•°ç‚¹ä»¥ä¸‹8000å…†æ¡ã‚ã€ã‚’é”æˆã—ãŸã¨ç™ºè¡¨ã—ãŸã€‚ ã—ã‹ã‚‚ãã®è¨ˆç®—ã‚’CPUã§ã¯ãªãï¼Œã¾ã—ã¦ã¯HPCï¼ˆHigh Performance Computingï¼‰å‘ã‘GPUã§ã‚ã‚‹Teslaã§ã‚‚ãªãï¼Œã”ãã”ãä¸€èˆ¬çš„ãªGeForceè£½å“ã§é”æˆã—ãŸã¨ã„ã†ã®ã ã‹ã‚‰é©šãã ã€‚ GPU Techno logy Conference 2013ï¼ˆGTC 2013ï¼‰ã§ã¯ï¼Œæœ¬ç ”ç©¶ã‚°ãƒ«ãƒ¼ãƒ—ã®ãƒªãƒ¼ãƒ€ãƒ¼ã§ã‚ã‚‹Ed Karrelsæ°ã«ã‚ˆã‚‹ã‚»ãƒƒã‚·ãƒ§ãƒ³ã€ŒComputing the Quadrillionth D
pochi-mk 2013/03/25
GPU

GPGPU

ã‚¤ãƒ™ãƒ³ãƒˆ

NVIDIA

ãƒ¬ãƒãƒ¼ãƒˆ

ç ”ç©¶
ãƒªãƒ³ã‚¯
Cacoo - Webä¸Šã§å›³ã®ä½œæˆã¨ãƒªã‚¢ãƒ«ã‚¿ã‚¤ãƒ ã‚³ãƒ©ãƒœãƒ¬ãƒ¼ã‚·ãƒ§ãƒ³
Online diagramming tool for collaborating on wireframes, flowcharts, and more
pochi-mk 2011/04/28
ãƒªãƒ³ã‚¯
1