[Info-vax] Re; Spiralog, RMS Journaling (was Re: FREESPADRIFT)

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax-bounces at info-vax.com] On Behalf Of
> Paul Sture via Info-vax
> Sent: 18-Jun-16 7:52 AM
> To: info-vax at info-vax.com
> Cc: Paul Sture <nospam at sture.ch>
> Subject: Re: [New Info-vax] Re; Spiralog, RMS Journaling (was Re:
> FREESPADRIFT)
> 
> On 2016-06-18, lawrencedo99 at gmail.com <lawrencedo99 at gmail.com>
> wrote:
> > On Saturday, June 18, 2016 at 8:56:38 PM UTC+12, Paul Sture wrote:
> >
> >> Rather useful when determining how much space is required for a
> >> Time Machine Backup, I imagine.
> >
> > Time Machine plays fast and loose with your filesystem integrity
> > (multiple hard links to directories), and you pay the price in
> > reliability
> >
> > <http://arstechnica.com/civis/viewtopic.php?t=1252811>.
> 
> Yes, I also went through problems with TM, but they appear to be
fixed.
> 
> TM can be terribly slow in certain circumstances.  I take regular
> clone copies using other software.
> 
> >> cf VMS only storing the block sizes of files rather than the number
of
> >> bytes; this can be problematic for utilities ported from a system
which
> >> has an easy/cheap way to supply the exact size in bytes.
> >
> > This was the example Linus Torvalds gave of why he hated VMS...
> 
> Eh?  There are many reasons to dislike an OS but 'hating VMS' for that
> *perceived* problem?  Really?
> 
> From another perspective, I ran straight into the spat between
Torvalds
> and Ted Ts'o when ext4 became the default on Ubuntu - this one did
> produce data loss for me (until I worked out what was going on):
> 
> "Linus Torvalds Upset over Ext3 and Ext4"
> 
> <http://www.linux-magazine.com/Online/News/Linus-Torvalds-Upset-
> over-Ext3-and-Ext4>
> 

With OpenVMS and its default write-through strategies, data integrity 
has always been the primary driver. 

Yes, this strategy does put it at a disadvantage when doing benchmarks 
that *only* consider performance. 

As illustrated in this article, with other platforms, there was a
conscious 
decision to implement write back (or async via periodic time frames) 
strategies as a way to increase perf even if the designers acknowledge 
that things like crashes, power fails could potentially cause data
integrity 
issues.

One only has to power off a system in the middle of a benchmark on that
system to see what the effects could be in the real world.

With new disks like SSD and really big memory technologies emerging, one

needs to ask whether the difference in perf is really worth  the
potential for
data integrity issues?  Especially when one considers the considerable
pain 
one must go through to recover from some significant event like a system

crash or hardware error or power failure on a write back designed
system.

Perhaps we will see more strategies like mirroring a RAM disk with a SSD

disk so that reads come from the RAM disk, but writes also go to SSD?

Regards,

Kerry Main
Kerry dot main at starkgaming dot com