News: 0001489650

  ARM Give a man a fire and he's warm for a day, but set fire to him and he's warm for the rest of his life (Terry Pratchett, Jingo)

Linux Patches Posted For RAID0 Atomic Write Support

([Linux Storage] 5 Hours Ago RAID0 Atomic Write)


Linux 6.11 introduces [1]block atomic write support including for NVMe and SCSI devices . With a new set of patches posted this week, atomic write support is wired up for the RAID0 MD code.

Building off the block atomic write support being introduced in Linux 6.11, the "request for comments" patches this week get atomic write support enabled for the MD RAID0 code while RAID1 support is still being analyzed. This allows for writes to be done atomically and there are also patches in existence for making use of these atomic flags in the likes of the XFS file-system.

The atomic write support depends upon support from the underlying NVMe / SCSI device. The proposed patches take into consideration the underlying bottom devices and whether they can properly handle atomic writes.

John Garry of Oracle sent out the [2]RAID0 atomic write patches for those interested in learning more about this proposed functionality.



[1] https://www.phoronix.com/news/Linux-6.11-Block-Atomic-Writes

[2] https://lore.kernel.org/lkml/20240903150748.2179966-1-john.g.garry@oracle.com/


piorunz

jabl

microchip8

It is a very humbling experience to make a multimillion-dollar mistake, but it
is also very memorable. I vividly recall the night we decided how to organize
the actual writing of external specifications for OS/360. The manager of
architecture, the manager of control program implementation, and I were
threshing out the plan, schedule, and division of responsibilities.

The architecture manager had 10 good men. He asserted that they could write
the specifications and do it right. It would take ten months, three more
than the schedule allowed.

The control program manager had 150 men. He asserted that they could prepare
the specifications, with the architecture team coordinating; it would be
well-done and practical, and he could do it on schedule. Furthermore, if
the architecture team did it, his 150 men would sit twiddling their thumbs
for ten months.

To this the architecture manager responded that if I gave the control program
team the responsibility, the result would not in fact be on time, but would
also be three months late, and of much lower quality. I did, and it was. He
was right on both counts. Moreover, the lack of conceptual integrity made
the system far more costly to build and change, and I would estimate that it
added a year to debugging time.
-- Frederick Brooks Jr., "The Mythical Man Month"