[Info-vax] Itanium Poulson a game changer

[MG]

On 22-feb-2013 14:53, ultradwc at gmail.com wrote:
> HotHardware: The Hottest Technologies, Tested and Burned In

No pun intended, right?


> Intel's Itanium has spent the past year in an unwelcome spotlight.

Just last year?  Oh, right, I forgot...  Only HP and Oracle count,
the people that ever cared for Alpha don't and never did.


> The war between HP and Oracle over whether or not the latter had an
> obligation to support HP servers after publicly promising to do so
> dragged Intel's Itanium roadmap into the limelight. Ultimately, the
> judge found that Oracle had to live up to its contractual obligations
> and concluded that the case was brought for personal reasons, but the
> damage to HP was done.

Interesting way of putting it and looking at it, so it was more or less
a victory for Oracle then.


> It adds execution units and rebalances those units to favor server
> workloads over HPC and workstation capabilities. Its multi-threading
> capabilities have been overhauled and it uses faster QPI linkages
> between the CPUs.

The million dollar question: Can it outrun my low-power (65 W), Core
i7?


> Dependencies and cache misses aren't as detrimental on Poulson as they
> were on earlier Itanium processors like Tukwila.

Okay, great, but what are we going to run?  OpenVMS, HP-UX and
NonStop?  The rx2800 i4 doesn't even list VMS yet, only HP-UX.

I may not prefer it, but tons of people like to run Linux and
Linux IA-64 is for all intents and purposes dead and has been
so for almost a decade now.  Good luck reviving that.

Also, good luck getting all those to 'remigrate'(?) to VMS,
those who already fled to Linux and Windows (with or without
AXP emulation).


> Total L3 cache bandwidth is estimated to be 700GB/s or higher.

Finally something that sets it apart.  Okay, but, one problem:
How is it going to work out in conjunction with, say, DDR3 memory
and other parts that could never reach such theoretical speeds?


> Itanium's original design was driven by the philosophy that software
> compilers should do all the heavy lifting when it came to optimization,
> parallelization, and scheduling. The Intel and HP engineers that
> workedon the project didn't think out-of-order execution would scale
> particularly well on x86 processors and built what they believed would
> be a high-performance microarchitecture that would scale far more
> effectively. In hindsight, OOoE designs outperformed these expectations
> and left Itanium high and dry.

They forgot to mention that AMD came up with a little thing called
AMD64, which 'spawned' EM64T and that was yet another nail in the
coffin of IA-64...  (But, why spoil a nice story?)


> It's not clear if Itanium has a long-term future.

I'm sure potential buyers will find that prospect particularly lovely.


> HP is the CPU's only major customer, and HP has stated that Oracle's
> shenanigans and FUD-spreading did significant damage to its Itanium
> server business. Even taken with a grain of salt, however, the Itanium
> 9500's performance figures are impressive. This could be the chip that
> breaks free of the "Itanic" moniker and establishes IA-64 as a healthy
> competitor for IBM's POWER 7.

"Shenanigans"?  (Is that the best 'behind-kissing' 'journalism' HP's
money could buy?)

Good luck defeating IBM, HP!  IBM has most likely more in-house
knowledge about VMS than HP had in all years that they've owned
VMS combined.  IBM has been hiring VMS people for years.  Not that
HP ever cared about VMS, in the first place, but... just sayin'!

Please IBM, buy VMS and then --- for all I care --- port it over
to PPC.  (It's not like VMS never ran on a RISC before, although
of a different endianness.)  Yes, I should stop day dreaming...

  - MG