[B! storage] dannã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

dann id:dann

storageã«é–¢ã™ã‚‹dannã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (47)

${{author_name}}$

{{{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}}

å†…è£½ã‚ªãƒ–ã‚¸ã‚§ã‚¯ãƒˆã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚µãƒ¼ãƒã€Œb3ã€ã§ã‚³ã‚¹ãƒˆæœ€é©åŒ–ã‚’ç›®æŒ‡ã—ãŸè©± - Mirrativ Tech Blog
ã‚¤ãƒ³ãƒ•ãƒ©ã‚¹ãƒˆãƒªãƒ¼ãƒŸãƒ³ã‚°ãƒãƒ¼ãƒ ã®è¿‘è—¤ (@udzura) ã§ã™ã€‚ä»Šå›žã¯ã€ãƒŸãƒ©ãƒ†ã‚£ãƒ–ã§å†…è£½ã—ã¦ã„ã‚‹ã‚ªãƒ–ã‚¸ã‚§ã‚¯ãƒˆã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚µãƒ¼ãƒã€Œb3ã€ã®ç´¹ä»‹è¨˜äº‹ã‚’æ›¸ããŸã„ã¨æ€ã„ã¾ã™ã€‚ ä»Šå›žã®è¨˜äº‹ã¯ã€6æœˆã«Gopher Talkã¨ã„ã†ã‚¤ãƒ™ãƒ³ãƒˆã§ç™ºè¡¨ã—ãŸã€ŒGoè£½ãƒŸãƒ‰ãƒ«ã‚¦ã‚§ã‚¢ã‚’å®Ÿè·µæŠ•å…¥ã™ã‚‹ã«ã‚ãŸã‚Šã‚„ã£ãŸã“ã¨ã€ã‚’ãƒ™ãƒ¼ã‚¹ã«ã€å†…å®¹ã‚’è©³ç´°ã«ã—ãŸã‚Šç›´è¿‘ã®é–‹ç™ºçŠ¶æ³ã«åˆã‚ã›ã¦æ›´æ–°ã—ãŸã‚‚ã®ã§ã™ã€‚ä¸€éƒ¨å†…å®¹ã¯ã“ã®ç™ºè¡¨ã¨é‡è¤‡ã—ã¦ã„ã¾ã™ãŒã”äº†æ‰¿ãã ã•ã„ã€‚ ã‚ªãƒ–ã‚¸ã‚§ã‚¯ãƒˆã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚µãƒ¼ãƒã‚’å†…è£½ã—ãŸèƒŒæ™¯ 1. å¤§é‡ã‚ªãƒ–ã‚¸ã‚§ã‚¯ãƒˆã®æ“ä½œã‚„å¢—ãˆç¶šã‘ã‚‹è»¢é€é‡ã«å¯¾å¿œã—ãŸã„ 2. ä¸€å®šæœŸé–“ã—ã‹ãƒ•ã‚¡ã‚¤ãƒ«ã®ä¿æŒã‚’ã—ãªã„ 3. ã‚ªãƒ³ãƒ¡ãƒ¢ãƒª/SSD/HDDã‚’çµ„ã¿åˆã‚ã›ãŸãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ãŒã—ãŸã„ ã‚ªãƒ–ã‚¸ã‚§ã‚¯ãƒˆã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸b3ã®ç‰¹å¾´ S3 äº’æ›ã®åŸºæœ¬çš„ãªAPIã‚’å®Ÿè£… LSM-Tree indexï¼‹WALãªDB/ãƒžãƒ¼ã‚¸æ“ä½œã«å¯¾å¿œ I/O å¸¯åŸŸã‚’åˆ¶é™å¯èƒ½ éžåŒæœŸãƒ¬ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³
dann 2023/10/23
storage
ãƒªãƒ³ã‚¯
https://www.microsoft.com/en-us/research/uploads/prod/2023/03/RDMA_Experience_Paper_TR-1.pdf
dann 2023/04/01
rdma

microsoft

storage
ãƒªãƒ³ã‚¯
Cray ClusterStor E1000ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚·ã‚¹ãƒ†ãƒ | HPE æ—¥æœ¬ | OID1012842049
ãƒ‘ãƒ•ã‚©ãƒ¼ãƒžãƒ³ã‚¹ã®åŠ¹çŽ‡æ€§ Cray ClusterStor E1000ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚·ã‚¹ãƒ†ãƒ ã§ã¯ã€NVMe PCIe 4.0ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚³ãƒ³ãƒˆãƒãƒ¼ãƒ©ãƒ¼ãŒä½¿ç”¨ã•ã‚Œã¦ã„ã‚‹ãŸã‚ã€ä»–ã®ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚½ãƒªãƒ¥ãƒ¼ã‚·ãƒ§ãƒ³ã«æ¯”ã¹ã¦ã€ã‚ˆã‚Šé«˜ã„ãƒ‘ãƒ•ã‚©ãƒ¼ãƒžãƒ³ã‚¹ã‚’å¼•ãå‡ºã›ã¾ã™ã€‚ã“ã®ã“ã¨ã¯å¿…è¦ãªã‚¨ãƒ³ã‚¯ãƒãƒ¼ã‚¸ãƒ£ãƒ¼ã€ãƒ©ãƒƒã‚¯ã€æ¶ˆè²»é›»åŠ›ã€ãŠã‚ˆã³ãƒ•ãƒã‚¢ã‚¹ãƒšãƒ¼ã‚¹ã‚’å‰Šæ¸›ã§ãã‚‹ã“ã¨ã‚’æ„å‘³ã—ã¾ã™ã€‚ 2ãƒ©ãƒƒã‚¯ãƒ¦ãƒ‹ãƒƒãƒˆãƒ•ã‚©ãƒ¼ãƒ ãƒ•ã‚¡ã‚¯ã‚¿ãƒ¼å†…ã®24å°ã®NVMe Gen4 SSDã«ã‚ˆã‚Šã€ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ ã®èªã¿å–ã‚Šã¨æ›¸ãè¾¼ã¿ã§åˆè¨ˆæœ€å¤§80GB/ç§’ã¨55GB/ç§’ã®ã‚¹ãƒ«ãƒ¼ãƒ—ãƒƒãƒˆã‚’å®Ÿç¾ã™ã‚‹Scala ble Storage Unit â€“ Flash (SSU-F) ã‚’æè¼‰ã—ãŸCray ClusterStor E1000ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚·ã‚¹ãƒ†ãƒ ãªã‚‰ã€ã‚ãšã‹24å°ã®SSDã‹ã‚‰ãŠä½¿ã„ã®ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ãƒˆãƒŽãƒ¼ãƒ‰ã«éžå¸¸ã«åŠ¹çŽ‡çš„ãªæ–¹æ³•ã§80GB/ç§’ã¨50GB/ç§’ã®ãƒ‘ãƒ•ã‚©ãƒ¼ãƒžãƒ³ã‚¹ã‚’ã‚‚ãŸã‚‰ã™ã“
dann 2023/02/18
storage
ãƒªãƒ³ã‚¯
å…±ç”¨ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸[æ±äº¬å¤§å¦æƒ…å ±åŸºç›¤ã‚»ãƒ³ã‚¿ãƒ¼ã‚¹ãƒ¼ãƒ‘ãƒ¼ã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ãƒ†ã‚£ãƒ³ã‚°éƒ¨é–€]
æ¦‚è¦ HPCI ã®è³‡æºã¨ã—ã¦ã€ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸è³‡æºã‚’ãƒ¦ãƒ¼ã‚¶ãƒ¼ã«æä¾›ã™ã‚‹ã‚‚ã®ã§ã™ã€‚ å…±ç”¨ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã«ã¯åºƒåŸŸåˆ†æ•£ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ Gfarm ã‚’åˆ©ç”¨ã—ã¦ã„ã¾ã™ gfarm2fs ã‚³ãƒžãƒ³ãƒ‰ã‚’ã¯ã˜ã‚ã¨ã—ã¦ã€Gfarm ã‚³ãƒžãƒ³ãƒ‰ãŒåˆ©ç”¨ã§ãã‚‹ç’°å¢ƒãŒãŠæ‰‹å…ƒã«ã‚ã‚Œã°ãã¡ã‚‰ã«å…±ç”¨ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã®ãƒ‡ãƒ¼ã‚¿é ˜åŸŸã‚’ãƒžã‚¦ãƒ³ãƒˆã—ã¦ã”ä½¿ç”¨é ‚ã‘ã¾ã™ã€‚ ã¾ãŸã€æ‰‹å…ƒã«ãã®ã‚ˆã†ãªç’°å¢ƒãŒãªã„æ–¹å‘ã‘ã« Gfarm ã‚³ãƒžãƒ³ãƒ‰ã‚’åˆ©ç”¨ã§ãã‚‹ã‚¯ãƒ©ã‚¤ã‚¢ãƒ³ãƒˆã‚‚ç”¨æ„ã—ã¦ã‚ã‚Šã¾ã™ã€‚ å¤§å®¹é‡ Gfarm ã¯è¤‡æ•°ã®ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ ãƒŽãƒ¼ãƒ‰ã‚’å˜ä¸€ã®ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ ã®ã‚ˆã†ã«ã¿ã›ã‚‹å®Ÿè£…ã§ã‚ã‚Šã€å¤§å®¹é‡ã®ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ ã‚’å®Ÿç¾ã™ã‚‹ã“ã¨ãŒå¯èƒ½ã§ã™ã€‚ ç¾åœ¨ã€å…±ç”¨ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ãŒãƒ¦ãƒ¼ã‚¶ãƒ¼å…¨ä½“ã§åˆ©ç”¨å¯èƒ½ãªç·å®¹é‡ã¯ 21.5 PB ã§ã‚ã‚Šã€ãƒ¦ãƒ¼ã‚¶ãƒ¼å€‹åˆ¥ã«å¯¾ã™ã‚‹ä½¿ç”¨å®¹é‡ã®åˆ¶é™ã¯ã€è¨ã‘ã‚‰ã‚Œã¦ã„ã¾ã›ã‚“ã€‚ å†—é•·åŒ– Gfarm ã¯è¤‡æ•°ã®ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ ãƒŽãƒ¼ãƒ‰ã‚’å˜ä¸€ã®ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ ã®ã‚ˆã†ã«ã¿ã›ã‚‹å®Ÿ
dann 2020/08/05
storage

server

backup
ãƒªãƒ³ã‚¯
https://www.t3.gsic.titech.ac.jp/sites/upload/TSUBAME3_Lustre_Seminar.pdf
dann 2020/07/01
lustre

storage
ãƒªãƒ³ã‚¯
Preferred Networks ã«ãŠã‘ã‚‹Hadoop - Preferred Networks Research & Development
Preferred Networks ï¼ˆä»¥ä¸‹PFNï¼‰ã§ã¯ã€ã€Œç¾å®Ÿä¸–ç•Œã‚’è¨ˆç®—å¯èƒ½ã«ã™ã‚‹ã€ã€Œå…¨ã¦ã®ã²ã¨ã«ãƒãƒœãƒƒãƒˆã‚’ã€ã¨ã„ã†ç›®æ¨™ã®ã‚‚ã¨ã€æ©Ÿæ¢°å¦ç¿’ã‚’å§‹ã‚ã¨ã—ãŸã‚ã‚‰ã‚†ã‚‹è¨ˆç®—æŠ€è¡“ã‚’ç”¨ã„ã¦ç ”ç©¶é–‹ç™ºã«æ—¥ã€…å–ã‚Šçµ„ã‚“ã§ã„ã¾ã™ã€‚ãã®éŽç¨‹ã§ã¯å¿…ãšã¨ã„ã£ã¦ã„ã„ã»ã©ã€ãƒ‡ãƒ¼ã‚¿ã®ä¿å˜ã‚„èªã¿å‡ºã—ãŒå¿…è¦ã«ãªã‚Šã¾ã™ã€‚ã“ã“ã§ã¯ã€æˆ‘ã€…ãŒã©ã®ã‚ˆã†ã«ãƒ‡ãƒ¼ã‚¿ç®¡ç†ã‚’ã—ã¦ã„ã‚‹ã‹ã€ã¾ãŸã€ãã®éŽç¨‹ã§ã©ã®ã‚ˆã†ã«Hadoopã‚’åˆ©ç”¨ã—ã¦ã„ã‚‹ã‹ã«ã¤ã„ã¦ç´¹ä»‹ã—ãŸã„ã¨æ€ã„ã¾ã™ã€‚ å†™çœŸ: PFNã‚«ãƒ©ãƒ¼ãƒªãƒ³ã‚°ã®ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚µãƒ¼ãƒãƒ¼ Hadoopå°Žå…¥ã®çµŒç·¯ HadoopãŒå¤šãåˆ©ç”¨ã•ã‚Œã¦ã„ã‚‹ã‚ˆã†ãªãƒã‚°åˆ†æžã‚„ã€ã‚¨ãƒ³ã‚¿ãƒ¼ãƒ—ãƒ©ã‚¤ã‚ºå‘ã‘ã®ETLå‡¦ç†ã‚„ãƒ‡ãƒ¼ã‚¿ã‚µã‚¤ã‚¨ãƒ³ã‚¹ã«æ¯”ã¹ã‚‹ã¨ã€æ·±å±¤å¦ç¿’ã®åˆ†é‡Žã§ã‚ˆãåˆ©ç”¨ã•ã‚Œã¦ã„ã‚‹ã‚ˆã†ãªè¦æ¨¡ã®ãƒ‡ãƒ¼ã‚¿ã¯æ¯”è¼ƒçš„å°ã•ã„ã§ã™ã€‚ã‚ˆãç”»åƒèªè˜ã®ãƒ™ãƒ³ãƒãƒžãƒ¼ã‚¯ã¨ã•ã‚Œã‚‹ImageNetã®2012å¹´ã®ã‚³ãƒ³ãƒšã§åˆ©ç”¨ã•ã‚ŒãŸãƒ‡ãƒ¼ã‚¿ã‚»ãƒƒãƒˆã¯200GBç¨‹åº¦ã§ã™ã€‚ã“ã‚Œã¯ãã‚‚ã
dann 2020/06/30
pfn

storage

hadoop
ãƒªãƒ³ã‚¯
E8 Storage Mellanox Solution Brief
dann 2019/08/27
nvmeof

mellanox

storage
ãƒªãƒ³ã‚¯
é«˜é€Ÿãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ ãƒ™ãƒ³ãƒãƒžãƒ¼ã‚¯ - ãƒ•ã‚¡ãƒŠãƒ†ã‚£ãƒƒã‚¯
é«˜é€Ÿãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã€€ãƒ™ãƒ³ãƒãƒžãƒ¼ã‚¯ 2016.07.12ã€€ãƒ™ãƒ³ãƒãƒžãƒ¼ã‚¯ãƒ¬ãƒãƒ¼ãƒˆ Infinibandã®æ€§èƒ½å‘ä¸Šã‚„Intel Omni-pathã®ç™»å ´ã«ã‚ˆã‚Šã€ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã®å¸¯åŸŸãŒã¤ã„ã«100Gbpsã«é”ã—ã¾ã—ãŸã€‚ã“ã®é«˜é€Ÿãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚’ä¸€éƒ¨ã®HPCã‚¹ãƒ‘ã‚³ãƒ³ç”¨ã ã‘ã«ä½¿ã£ã¦ã„ã‚‹ã®ã¯ã‚‚ã£ãŸã„ãªã„ã¨ã¯æ€ã„ã¾ã›ã‚“ã‹ï¼Ÿ ã“ã‚Œã‚‰ã‚’ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚·ã‚¹ãƒ†ãƒ ã«å°Žå…¥ã™ã‚Œã°ã€FC-SANç‰ã«æ¯”ã¹ã¦åœ§å€’çš„ã«é«˜ã‚³ã‚¹ãƒˆãƒ‘ãƒ•ã‚©ãƒ¼ãƒžãƒ³ã‚¹ã®SAN/NASç’°å¢ƒãŒæ§‹ç¯‰å¯èƒ½ã§ã™ã€‚ã•ã‚‰ã«RDMAæ©Ÿèƒ½ã‚’åˆ©ç”¨ã™ã‚Œã°ã€è¶…ä½Žãƒ¬ã‚¤ãƒ†ãƒ³ã‚·ã‚¢ã‚¯ã‚»ã‚¹ã«ã‚ˆã‚Šã•ã‚‰ãªã‚‹æ€§èƒ½å‘ä¸Šã‚‚æœŸå¾…ã§ãã¾ã™ã€‚ï¼ˆå®Ÿéš›ã«ä¸€éƒ¨å¤§æ‰‹ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ãƒ¡ãƒ¼ã‚«ãƒ¼è£½å“ã§ã‚‚ã“ã®æŠ€è¡“ã¯æŽ¡ç”¨ã•ã‚Œã¦ã„ã¾ã™ã€‚ï¼‰ RDMAã¨ã¯â€¦ Remote Direct Memory Accessã®ç•¥ã€‚ã‚µãƒ¼ãƒé–“ã®ãƒ‡ãƒ¼ã‚¿è»¢é€ã‚’ã€CPUã‚’ä»‹ã•ãšã«ç›´æŽ¥ãƒ¡ãƒ¢ãƒªåŒå£«ã§è¡Œã†æŠ€è¡“ã€‚ ã¾ãŸã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã‚’é«˜é€ŸåŒ–ã™ã‚‹æ–¹æ³•ã¨ã—ã¦ã€è¿‘å¹´ã§ã¯Fl
dann 2018/12/15
storage
ãƒªãƒ³ã‚¯
February 2016 Webinar Series - Architectural Patterns for Big Data on AWS
With an ever-increasing set of techno logies to process big data, organizations often struggle to understand how to build scala ble and cost-effective big data applications. In this webinar, we will simplify big data processing as a pipeline comprising various stages; and then show you how to choose the right techno logy for each stage based on criteria such as data structure, design patterns, and b
dann 2016/12/10
storage
ãƒªãƒ³ã‚¯
http://tce.technion.ac.il/wp-content/uploads/sites/8/2015/06/SC-1.3-A.Kesselman.pdf
dann 2015/07/12
google

storage
ãƒªãƒ³ã‚¯
F4 - Photo Storage at Facebook
Featuring: Joe Gasperetti, Production Engineer at Facebook;Â and Satadru Pan, Software Engineer at Facebook Description: none
dann 2014/11/16
facebook

storage
ãƒªãƒ³ã‚¯
Ceph ã®ã‚¹ãƒˆãƒãƒ¼ã‚¢ãƒ«ã‚´ãƒªã‚ºãƒ (straw algorithm) - NAKAMURA Minoru's Diary
2002 | 10 | 11 | 12 2003 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 2004 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 2005 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 2006 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 2007 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 2008 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 2009 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 1
dann 2014/01/03
algorithm

ceph

storage
ãƒªãƒ³ã‚¯
WalB v1.0 ãƒªãƒªãƒ¼ã‚¹ - Cybozu Inside Out | ã‚µã‚¤ãƒœã‚¦ã‚ºã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ã®ãƒ–ãƒã‚°
æ›´æ–°æƒ…å ± 2017å¹´6æœˆç¾åœ¨ã€WalB ã®ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆãƒšãƒ¼ã‚¸ã¯https://walb-linux.github.io/ã§ã™ã€‚ ã“ã‚“ã«ã¡ã¯ï¼Žã‚µã‚¤ãƒœã‚¦ã‚ºãƒ»ãƒ©ãƒœã®æ˜Ÿé‡Žã§ã™ï¼Ž ä»Šæ—¥ã¯ï¼ŒåŠ¹çŽ‡çš„ãªãƒãƒƒã‚¯ã‚¢ãƒƒãƒ—ã¨éžåŒæœŸãƒ¬ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã‚’ç›®çš„ã¨ã—ãŸ Linux ã‚«ãƒ¼ãƒãƒ«ãƒ‰ãƒ©ã‚¤ãƒ WalB ã«ã¤ã„ã¦ã®ç°¡å˜ãªç´¹ä»‹ã¨ï¼Œãã®ãƒãƒ¼ã‚¸ãƒ§ãƒ³ 1.0 ã®ãƒªãƒªãƒ¼ã‚¹å‘ŠçŸ¥ã‚’ã•ã›ã¦é ‚ãã¾ã™ï¼Ž WalB ã£ã¦ä½•ï¼Ÿ WalB ã¯ã€Œãƒ¯ãƒ«ãƒ“ãƒ¼ã€ã¨ç™ºéŸ³ã—ã¾ã™ï¼Žåå‰ã®ç”±æ¥ã¯ Block-level WAL ã§ã™ï¼ŽWAL ã¯ Write-Ahead Logging (ãƒã‚°å…ˆè¡Œæ›¸ãè¾¼ã¿) ã®ã“ã¨ã§ï¼Œãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ã®ä¸€è²«æ€§åˆ¶å¾¡ç‰ã«ç”¨ã„ã‚‰ã‚Œã¾ã™ï¼Ž WalB ã‚’ä½¿ã†ã¨ï¼Œã‚ªãƒ³ãƒ©ã‚¤ãƒ³çŠ¶æ…‹ã®ãƒ–ãƒãƒƒã‚¯ãƒ‡ãƒã‚¤ã‚¹ã«ãŠã„ã¦ï¼Œã‚¢ãƒ—ãƒªã‚±ãƒ¼ã‚·ãƒ§ãƒ³ã‚„ãƒ•ã‚¡ã‚¤ãƒ«ã‚·ã‚¹ãƒ†ãƒ ã«ã‚ˆã‚‹ãƒ‡ãƒ¼ã‚¿æ›¸ãè¾¼ã¿ã«ã‚ˆã£ã¦ç”Ÿã˜ã‚‹å·®åˆ†ãƒ‡ãƒ¼ã‚¿ã‚’ï¼Œãƒ–ãƒãƒƒã‚¯ãƒ‡ãƒã‚¤ã‚¹ã‚’ãƒ•ãƒ«ã‚¹ã‚ãƒ£ãƒ³ã™ã‚‹ã“ã¨ãªãåŠ¹çŽ‡çš„ã«å–å¾—å‡ºæ¥ã‚‹
dann 2013/05/29
linux

storage
ãƒªãƒ³ã‚¯
An Efficient Backup and Replication of Storage
This document describes WalB, a Linux kernel device driver that provides efficient backup and replication of storage using block-level write-ahead logging (WAL). It has negligible performance overhead and avoids issues like fragmentation. WalB works by wrapping a block device and writing redo logs to a separate log device. It then extracts diffs for backup/replication. The document discusses WalB'
dann 2013/05/29
backup

storage
ãƒªãƒ³ã‚¯
International Parallel Data Systems Workshop
Program Co-Chairs: The Ohio State University, USA Illinois Institute of Techno logy, USA Reproducibility Co-Chairs: Lawerence Berkeley National Laboratory, USA RWTH Aachen University, Germany abstract / cfp / submissions / WIP session / workshop registration / committees PDSW24 Reproducability Addendum SUBMISSION DEADLINE EXTENDED: AUG 9, 2024 - final deadline Invited speaker: TBA agenda Agenda inf
dann 2013/01/01
hpc

storage
ãƒªãƒ³ã‚¯
ã‚¯ãƒ©ã‚¦ãƒ‰ã‚¤ãƒ³ãƒ•ãƒ©ã‚’æ”¯ãˆã‚‹é«˜æ€§èƒ½ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã€ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸æŠ€è¡“
ã‚«ãƒ†ã‚´ãƒªãƒ¼ DX (2) ä¸€èˆ¬ (58) ç ”ç©¶ä¼š (6) åƒãæ–¹ (4) æŠ€è¡“ (349) Edge AI (2) Edge Computing (12) Erlang (1) FIWARE (2) Fog Computing (9) Infiniband (31) Internet of Things (32) Key Value Store (17) Linux (3) Linux KVM (10) Machine Learning (2) RealTime Web (14) Webã‚µãƒ¼ãƒ“ã‚¹ (42) ã‚¤ãƒ³ãƒ•ãƒ© (7) ã‚³ãƒ³ãƒ†ãƒŠ (3) ã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ (92) ãƒ‡ãƒ¼ã‚¿ã‚»ãƒ³ã‚¿ãƒ¼ (7) ãƒ‡ãƒ¼ã‚¿ãƒ™ãƒ¼ã‚¹ (47) ãƒ‡ãƒ¼ã‚¿æµé€š (6) ãƒ†ãƒ¬ãƒ—ãƒ¬ã‚¼ãƒ³ã‚¹ (2) ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ (214) ä»®æƒ³åŒ– (110) ç½å®³ã‚³ãƒŸãƒ¥ãƒ‹ã‚±ãƒ¼ã‚·ãƒ§ãƒ³ (26) ç©ºé–“æƒ…å ± (30) é‡åã‚³ãƒ³ãƒ”ãƒ¥ãƒ¼ãƒ†ã‚£ãƒ³ã‚° (3) æ•™è‚²
dann 2012/11/25
network

storage
ãƒªãƒ³ã‚¯
OpenStack Storage(Swift) èª¿æŸ»å ±å‘Šæ›¸ - ã‚¯ãƒªã‚¨ãƒ¼ã‚·ãƒ§ãƒ³ãƒ©ã‚¤ãƒ³æ ªå¼ä¼šç¤¾
2011å¹´1æœˆ18æ—¥ 1. æ¦‚è¦ Swift ã¯ OpenStack ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆãŒé–‹ç™ºã—ã¦ã„ã‚‹åˆ†æ•£ã‚ªãƒ–ã‚¸ã‚§ã‚¯ãƒˆã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã§ã‚ã‚‹ã€‚OpenStack ã¯ã€ã‚¯ãƒ©ã‚¦ãƒ‰ç’°å¢ƒã‚’æ§‹ç¯‰ã™ã‚‹ãŸã‚ã®ã‚ªãƒ¼ãƒ—ãƒ³ã‚½ãƒ¼ã‚¹ãƒ—ãƒã‚¸ã‚§ã‚¯ãƒˆã§ã‚ã‚Šã€ä»®æƒ³ãƒžã‚·ãƒ³ã®ç®¡ç†ã‚’è¡Œã† Nova ã¨ã€ã‚ªãƒ–ã‚¸ã‚§ã‚¯ãƒˆã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã® Swift ãŒã‚ã‚‹ã€‚Swift ã®ä½ç½®ã¥ã‘ã¨ã—ã¦ã¯ã€Nova ã®ä»®æƒ³ãƒžã‚·ãƒ³ ã®ã‚¤ãƒ¡ãƒ¼ã‚¸ã®ä¿å˜ãªã©ãŒæŒ™ã’ã‚‰ã‚Œã‚‹ã€‚ã—ã‹ã—ã‚ªãƒ–ã‚¸ã‚§ã‚¯ãƒˆã‚¹ãƒˆãƒ¬ãƒ¼ã‚¸ã®æ©Ÿèƒ½ã¯ç‹¬ç«‹ã—ã¦ãŠã‚Šã€Nova ã¨ é–¢ä¿‚ãªãåˆ©ç”¨ã™ã‚‹ã“ã¨ã‚‚å¯èƒ½ã§ã‚ã‚‹ã€‚ç¾è¡Œã®ä»•æ§˜ã§ã¯ã€Nova ã¨é€£æºã™ã‚‹ã‚ˆã†ã«ã¯ãªã£ã¦ã„ãªã„ãŸã‚ã€ æœ¬å ±å‘Šæ›¸ã§ã¯ Swift ã®ã¿ã®åˆ©ç”¨ã‚’ä¸å¿ƒã«è¿°ã¹ã‚‹ã€‚ Swift ã®å…ƒã«ãªã£ãŸã®ã¯ã€Rackspace ç¤¾ã®é‹ç”¨ã™ã‚‹ CloudFiles ã¨ã„ã†ã‚µãƒ¼ãƒ“ã‚¹ã§ã‚ã£ãŸã€‚ç¾åœ¨ ã¯ã‚ªãƒ¼ãƒ—ãƒ³ã‚½ãƒ¼ã‚¹ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ã¨ã—ã¦å…¬é–‹ã•ã‚Œã¦ãŠã‚Šã€é–‹ç™ºè¨€èªžã«ã¯ Python
dann 2012/06/01
architecture

storage

distributed
ãƒªãƒ³ã‚¯
MythbustingFlash_081011v1.ppt
dann 2012/03/05
oracle

storage

cool
ãƒªãƒ³ã‚¯
CRUSH: Controlled, Scalable, Decentralized Placement of Replicated Data [PDF]
CRUSH: Controlled, Scala ble, Decentralized Placement of Replicated Data Appeared in Proceedings of SC '06. Abstract Emerging large-scale distributed storage systems are faced with the task of distributing petabytes of data among tens or hundreds of thousands of storage devices. Such systems must evenly distribute data and workload to efficiently utilize available resources and maximize system perf
dann 2012/02/24
storage

datadistribution
ãƒªãƒ³ã‚¯
ReadyNASã¨ã¯ä½•ã‹
ReadyNASã‚’æ™®é€šã«ä½¿ã£ã¦ã„ã‚‹é™ã‚Šã¯å…¨ãæ°—ã«ã™ã‚‹å¿…è¦ã¯ãªã„ãŒã€ä½•ã‹å•é¡ŒãŒèµ·ããŸã‚Šã—ãŸã¨ãã«ã€ãã‚‚ãã‚‚ReadyNASã®ä¸ã¯ã©ã†ãªã£ã¦ã‚‹ã®ã‹ã¨ã„ã†ã“ã¨ã«èˆˆå‘³ãŒæ¹§ãã®ã¯è‡ªç„¶ã ã¨æ€ã†ã®ã§ã€ãã®æµã‚Œã®è©±ã€‚ ï¼ˆã“ã®ã‚¨ãƒ³ãƒˆãƒªã¯ä¸€ç¶šãã®ã‚¨ãƒ³ãƒˆãƒªã®5/5ï¼‰ 4.1. ReadyNASã®æ§‹æˆï¼ˆæŽ¨æ¸¬ï¼‰ ReadyNASãŒä½•ã‹ã¨ã„ãˆã°ã€Linuxã‚’OSã¨ã—ã¦ç¨¼åƒã™ã‚‹/RAIDæ§‹æˆã§/NASå°‚ç”¨ã®PCã€ã¨ã„ã†ã“ã¨ã«ãªã‚‹ã€‚ã‚¢ãƒ—ãƒ©ã‚¤ã‚¢ãƒ³ã‚¹ã®Linuxæ©Ÿã¨ã—ã¦ã¯æ™®é€šã®PCã«è¿‘ã„ãŒã€æ™®é€šã®PCã¨ç‰¹ã«é•ã†ã®ã¯ã€ ã‚ªãƒ³ãƒœãƒ¼ãƒ‰ã«OSã®ã‚¤ãƒ³ã‚¹ãƒˆãƒ¼ãƒ«ã‚¤ãƒ¡ãƒ¼ã‚¸ã‚’æŒã£ã¦ã„ã‚‹ã“ã¨ ãƒ–ãƒ¼ãƒˆãƒ—ãƒã‚»ã‚¹ãŒå°‘ã—å¤‰ã‚ã£ã¦ã„ã‚‹ã“ã¨ ãŸã ã€Netgearã®äººã¯ã“ã†ã„ã†è©±ã‚’ã‚ã¾ã‚Šæ£é¢ã‹ã‚‰æ›¸ã„ã¦ãã‚Œã¦ãªã„ã€‚ã‚ãã¾ã§ã‚¢ãƒ—ãƒ©ã‚¤ã‚¢ãƒ³ã‚¹ã§ã‚ã£ã¦ãƒ¦ãƒ¼ã‚¶ãƒ¼ã®å¿ƒé…ã™ã‚‹ã“ã¨ã§ã¯ãªã„ã—ã€SSHã§ãƒã‚°ã‚¤ãƒ³ã™ã‚Œã°åˆ†ã‹ã‚‹äººã«ã¯åˆ†ã‹ã‚‹ã ã‚ã†ã€ã¨èª¬æ˜Žã®å¿…è¦ã‚’æ„Ÿã˜ã¦ãªã„ã®ã‹ã‚‚ã—ã‚Œãª
dann 2012/02/12
storage
ãƒªãƒ³ã‚¯
1 2 3 æ¬¡ã®ãƒšãƒ¼ã‚¸