[Info-vax] VMS - The new file system. What do we know about it?...

[Dirk Munk]

Michael Moroney wrote:
> IanD <iloveopenvms at gmail.com> writes:
>
>> I'm still curious about the new file system
>
>> What else can be let loose about it?
>
>> - Will existing RMS work with it?
>
> The idea is that existing RMS will work with it.  The new file system will
> have increased performance, and will work on disks larger than 2TB.
>
>> - Does it working across volumes? (i.e. sharding like abilities)
>
> Do you mean volume sets? According to the bootcamp presentation, the
> answer is no. Volume sets are a mostly obsolete concept anyway.
>
>> - I understand that it's being revamped to 64 bit internals, what are
>> the limits (i.e. How does it compare to this list/ What's similar on this
>> list?)
>
> I do not understand the question.  Right now, as Clair mentioned, work is
> being done to the executive internals so that disks larger than 2TB will
> work correctly. This should be in V8.next with the goal being that one can
> $ MOUNT/FOREIGN a >2TB drive and do logical $QIO reads and writes to any
> valid block on the drive, but not do RMS access. This is low level stuff,
> like "Give me Disk Block nnnn" or "Give me X blocks (actually X*512 bytes)

Modern disks of course use the Advanced Format structure, with 4kB 
blocks/sectors. Using arbitrary 512 byte blocks is possible, but not 
very good for performance. Is that taken into consideration?

> starting at Disk Block nnnn". It's more complicated than it sounds.
> Marrying the new file system to the 64 bit LBN stuff will have to wait for
> V9.x.
>