9 Apr
2013
9 Apr
'13
2:40 a.m.
On 9-apr-2013 2:35, Liam Proven wrote:
I suggest that you both retreat and attempt to clarify
your positions.
Feel free to point out where it has been vague and inflammatory,
because I don't at all feel addressed by such accusations.
"MG" - you seem now to be comparing
mainframes to DEC OpenVMS boxes,
is that correct?
To the degree that they're both niche platforms, and nowadays more
than they ever were before, yes. Not architecturally.
As I already said, I have never even gotten the opportunity to use
a mainframe, other than perhaps what one can do through emulation
(e.g. through SimH) with various historical offerings.
The closest thing to it, the 'quasi-mainframe' as I dared to call
it, was the public access AS/400 of Rechenzentrum Kreuznach. The
only "i" I have ever seen and used (only as an unprivileged user
at that.)
Are you asking why mainframes are still around while
DEC's OpenVMS
offerings are long gone?
Because VMS is supposedly also alive, like the mainframe. But,
at least many VMS people are a bit more honest with themselves
on average and show a bit more self-criticism than the average
IBM (and especially mainframe) type I've been coming across in
the last few years.
Or are you pointing out that, toward the end, OpenVMS
boxes morphed
into something not unlike high-end PCs and asking why mainframes
have not done the same?
No and I'm not sure how you got this out of all that came by so
far. In fact, this was never (from the beginning) VMS' strong
point.
VAXes and their kin - even big ones - were not true
mainframes.
I guess not, but then, the definition of "mainframe" is not one
I care tremendously for. I never lost sleep over whether or not
they were considered a "mainframe".
There used to be 3 types of computer: mainframes,
minicomputers and
microcomputers. At the high end, micros blended into the specialised
realm of "workstations". (Obviously these are sweeping generalisations
here.)
This is, of course, a bit of an IBM-dictated 'taxonomy'. I mean,
DEC didn't even call its systems "computers" literally at first.
(Think of "PDP".)
Would that mean a "PDP-11" isn't a computer therefore...?
The differential between them was this:
* micros run off a microprocessor, a single-chip CPU, and were
essentially designed for a single, interactive user
* minis ran off CPUs built out of discrete parts - pre-microchip - and
were designed to serve a small number of interactive users on
terminals
* mainframes predate the whole notion of interactive users and aren't
really designed to serve interactive, logged-in users at all; instead,
they were designed and very heavily optimised for running batch jobs
with great efficiency and reliability.
Yes, I'm aware of the notion and idea of "time sharing". Very ancient
stuff though, all in all, to be honest.
Mainframes are a conceptually different type of
computer. They don't
have keyboards and mice, obviously, unlike workstations; they don't
even support conventional terminals, i.e. dumb terminals running over
serial ports.
I may not have used a mainframe, as I said, but I'm not /that/ unaware
about their functioning.
Many of these characteristics, like with regard to dumb terminals,
are also true for "i" though. At least, I can't think of an "i"
(or, AS/400) that would be operated via direct graphics head/frame-
buffer interface with a 'keyboard & mouse'; you?
That is historical now, but the point is, mainframes
are not
interactive computers, and that's why they've survived and that's
also why they never mutated into workstations as the VAX and Alpha
did.
It's funny you should mention that, but they were rather poor for
those purposes overall (eventually in the long run).
Various things.
* Specialist OSs, so that, for instance, unlike with PC or Unix
virtualisation, the hypervisor OS is nothing but a hypervisor, whereas
the guest often depends on a hypervisor for its function - some guest
OSs don't even have things like networking or filesystems, because the
host provides this. Sounds weird but it's tens to hundreds of times
more efficient than the PC model.
I'm aware of some of these concepts. Say, isn't this what the FreeBSD
"jails" are slowly, but surely, trying to mimic a bit?
* Everything is offloaded. These systems have multiple
processors,
sure, like a high-end server, but they have lots of different types of
processor. Some do computation, some do encryption, some manage
databases, some just do indexing, some just handle various different
types of I/O. PC servers cannot even come close to this, but the PC's
efforts at comparing are things like machines with TCP/IP offload
engines in their network cards, stacks of dozens of GPGPU cards for
number crunching, and *in the same case as the PC* both NAS storage
and SAN storage, talking iSCSI to some devices, Fibrechannel to
others, SMB to others, NFS to others - all inside a single system,
using whatever is more appropriate for each workload. Smart dedicated
sub-computers running each type of storage, so that the "main"
processor doesn't do /any/ of this for itself; it /only/ runs the
all-important batch workloads.
Like I said, it's a beautiful sounding system, I never doubted that.
My main gripe is the 'elitism' that IBM seems to instill and (like
also someone else admitted) seems to artificially keep alive.
It also --- well, to me anyway --- gives the impression that it's
more of a money making scheme (the 'exclusivity', so to speak)
than a sound future-proof treatment.
Imagine a whole server room, hell, a whole datacentre,
with hundreds
of independent servers - some running Windows, some Linux, some
Solaris, some Netapp Filers, some dedicated SQL servers, all in a
single rack, managed as a single instance, with 100% compatibility and
all the components, from the processor chips to the disk drives to the
network cards to all the OSs, all coming from a single vendor, all
optimised for handling big server workloads with /better than/ 99.999%
availability.
I've never seen official performance statistics, just IBM's own
figures. So, I can't comment on how it truly behaves in this
regard.
That is why people still buy (or more to the point,
rent) mainframes.
There is a generation, I'm even willing to bet several generations,
that grew up with nothing other than Windows and Linux. Some young
enough have never even experienced nor seen/heard an IBM PC, let
alone the term "IBM PC".
Why is this important? Because IBM itself, the company, is also
falling further into obscurity like this, along with "z".
Will banks continue to run "z"? I guess some will, but I also
read about NonStop and, not surprisingly, the ever encroaching
Linux and even Windows.
Then there's also the question of the current and upcoming
generations, freshly indoctrinated with notions of "the cloud"
and what-not: How "cloud-ready" is "z"? And to what degree
would they prefer "z" over some Dell or 'brandless' x86 or even
ARM (those seem to be coming, too, now) server?
Because when it comes to the point when you are going
to have to spec
an entire datacentre, hire a whole team of experts to integrate it
all, and spent a few million a year running it, then in some cases, it
makes good financial sense to just lease a single box from IBM which
does all of this in one fridge-sized cabinet that sits there and just
works. No integration, no management, software compatibility that goes
back to whole decades before the 8086 was invented in 1978 or
whatever, all guaranteed and backed up by the most solid, high-quality
SLA that has ever existed in the IT industry.
Like I wrote before, in the case of the "GAMEframe" setup by that
Brazilian company Hoplon, they wrote they had to offload to "Cell"
processor 'blades'; because otherwise it'd 'tax the "z" too
much'.
When I thus read such things, I become somewhat doubtful of such
claims. It doesn't also help that IBM has made it so relegated
and secluded to themselves and their direct customers, that the
inquiring minds have little insight into these (and hopefully
truthful?) types of performance benchmarks and figures...
What IBM did in the last decade or so is realise that
this honking
great boxes can run Linux in one of their virtualization partitions
just as well as they can run weird proprietary IBM OSs. And if you run
Linux in that VM, then you get all the PC-type stuff that mainframes
don't do terribly well for free - TCP/IP, HTTP, all that sort of
stuff.
How does this work in combination with the "time sharing" operating
principles, though? Please bear with me, as I don't have any
direct experience myself.
But the scalability of a mainframe means that whereas
on a very
well-specced x86 server, you can run dozens of VMs, maybe even a
hundred plus if you set it up very carefully and throw terabytes of
RAM at it, on a bog-standard low-end mainframe, you can run tens of
*thousands* of Linux instances all at once - because running lots and
lots of similar workloads side-by-side and keeping them all responsive
is what mainframes are built to do.
Though, with terabytes of RAM, I think many other platforms would
also fare well...
I am not talking about a system that is 5? or 10? more
scalable.
I'm talking about something 50? or 100? more scalable. Not
supporting hundreds of users per box, but millions of users per
box.
Why isn't IBM more eager to speak of this and show the world what
"z" is truly capable of? Why isn't YouTube loaded with videos
showing these kind of things of, to name something?
I really don't get it, such a capable platform (I'm told), but
absolutely no desire to expand and increase its user/install
base?
Sure, only on certain specialised workloads, not on
pure CPU-intensive
stuff - but for finance and the like, stuff for which there is code
out there that has been in production since the 1960s, a level of
maturity that is literally impossible for x86 or Unix products.
I guess that's another reason why the "GAMEframe" used those
"Cell" processor 'blades'?
But if you want to serve files on a LAN, or run a
thousand instances
of MariaDB, Perl and Apache running some JSON queries and rendering
PHP, no, it's a stupid, ruinously expensive way to do that.
I'll gladly take your word for it, it sounds like you have more
experience with mainframes than I do.
But, what are your predictions for the future? I mean, it's exactly
these things that are ever-expanding and becoming more and more
common nowadays, aren't they?
- MG