> On 7 Jul 2016, at 02:46, Chris Murphy <lists@xxxxxxxxxxxxxxxxx> wrote: > Chaps, I didn’t wanted this to spring up as a performance of btrfs argument, BUT you are throwing a lot of useful data, maybe diverting some of it into wiki ? you know, us normal people might find it useful for making educated choice in some future ? :) Interestingly on my RAID10 with 6 disks I only get: dd if=/mnt/share/asdf of=/dev/zero bs=100M 113+1 records in 113+1 records out 11874643004 bytes (12 GB, 11 GiB) copied, 45.3123 s, 262 MB/s filefrag -v ext: logical_offset: physical_offset: length: expected: flags: 0: 0.. 2471: 2101940598..2101943069: 2472: 1: 2472.. 12583: 1938312686..1938322797: 10112: 2101943070: 2: 12584.. 12837: 1937534654..1937534907: 254: 1938322798: 3: 12838.. 12839: 1937534908..1937534909: 2: 4: 12840.. 34109: 1902954063..1902975332: 21270: 1937534910: 5: 34110.. 53671: 1900857931..1900877492: 19562: 1902975333: 6: 53672.. 54055: 1900877493..1900877876: 384: 7: 54056.. 54063: 1900877877..1900877884: 8: 8: 54064.. 98041: 1900877885..1900921862: 43978: 9: 98042.. 117671: 1900921863..1900941492: 19630: 10: 117672.. 118055: 1900941493..1900941876: 384: 11: 118056.. 161833: 1900941877..1900985654: 43778: 12: 161834.. 204013: 1900985655..1901027834: 42180: 13: 204014.. 214269: 1901027835..1901038090: 10256: 14: 214270.. 214401: 1901038091..1901038222: 132: 15: 214402.. 214407: 1901038223..1901038228: 6: 16: 214408.. 258089: 1901038229..1901081910: 43682: 17: 258090.. 300139: 1901081911..1901123960: 42050: 18: 300140.. 310559: 1901123961..1901134380: 10420: 19: 310560.. 310695: 1901134381..1901134516: 136: 20: 310696.. 354251: 1901134517..1901178072: 43556: 21: 354252.. 396389: 1901178073..1901220210: 42138: 22: 396390.. 406353: 1901220211..1901230174: 9964: 23: 406354.. 406515: 1901230175..1901230336: 162: 24: 406516.. 406519: 1901230337..1901230340: 4: 25: 406520.. 450115: 1901230341..1901273936: 43596: 26: 450116.. 492161: 1901273937..1901315982: 42046: 27: 492162.. 524199: 1901315983..1901348020: 32038: 28: 524200.. 535355: 1901348021..1901359176: 11156: 29: 535356.. 535591: 1901359177..1901359412: 236: 30: 535592.. 1315369: 1899830240..1900610017: 779778: 1901359413: 31: 1315370.. 1357435: 1901359413..1901401478: 42066: 1900610018: 32: 1357436.. 1368091: 1928101070..1928111725: 10656: 1901401479: 33: 1368092.. 1368231: 1928111726..1928111865: 140: 34: 1368232.. 2113959: 1899043808..1899789535: 745728: 1928111866: 35: 2113960.. 2899082: 1898257376..1899042498: 785123: 1899789536: last,elf If it would be possible to read from 6 disks at once maybe this performance would be better for linear read. Anyway this is a huge diversion from original question, so maybe we will end here ? -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
