* Posts by GrumpyOF

38 publicly visible posts • joined 9 Jan 2014

Lost in translation and adrift in cloud storage


With the lockdown, sorry, but no beer.

Open the pod bay doors: Voice of HAL 9000 Douglas Rain dies at 90


Re: Fun IT facts about HAL's song

Back in the '60s there was an IBM 1130 in the lab of the Electrical Engineering faculty at Liverpool Uni, that played YellowBird via RF. Probably not unique but it was kinda spooky back then.

Now IBM turns redundo gun on its Digital Business Group


Re: Not long now then....

I believe the "IBM of old" started disappearing round about 1984, under guidance of Cary, Opel and Akers

it was going to be a 100 billion $ company by 1990!! I actually heard them say it when they visited us down South. Think it was Opel at the time.

After their disastrous leadership they brought in an outsider (first time ever) and he started the merry-go--round of fire and hire that seems to have been IBM's mode of operation ever since. (other very large organisations seem to have copied them ad nauseum).

AI smarts: IBM pushes out 'faster than X86' POWER9 servers


POWER is good, if used correctly

Have been involved with a client who has run POWER for many years extremely successfully. They still have several POWER 7 795 machines on the floor (out of maintenance but running happily). In over six years there has been only 1 hardware issue (caused by an overzealous engineer who pulled a wrong cable).

They run POWER VM and AIX and that is where some issues arise---not the products themselves but rather the lack of suitable skills to maintain and optimise both the virtualisation firmware and the OS.

Unfortunately none of the support guys ever played on a mainframe with MDF, PR/SM or MLPF and virtualisation on x86 is trivial (asking for trouble there).

They could have gone Linux but converting/testing/migrating over 300 logical partitions running AIX was a step too far.

Have got a couple of big POWER 8 boxen on the floor now.

The capex for thes products always look horrific cf other platforms but if you do proper costing anaylses over 6 years and rising, with guarenteed performance and availability POWER certainly competes against other solutions financailly.

Phew! We had to order a new array to hold all of this week's Dell EMC announcements


Should be fun

watching the EMC sales reps mangling product positioning as they seem to have done for years.

Is this a deliberate attempt by Dell EMC to obfuscate the whole storage market place?, 'cos' it does seem to be a total product overload

Gang-briefed by IBM bosses in Hawaii? Nah, I'll take redundancy


Re: Clueless enough to be an IBM executive

It's not the people before. The problem goes back to the '80s with the like af Carey and Aikers (long time IBM veterans) who were forecasting that IBM would be a $100 Billion revenue company.

Boy ,did they get that wrong. It took until Gerstner (non IBM) came along, cut the work force by 40%----yep about 40%, before things started to clear up. Since then the company has expanded and contracted personnel numbers with gay abandon.

Years of poor management, poor reading of the market place and only some reasonable execution in very limited areas.

No quick fix for this

Stop us if you've heard this one before: IBM sheds more workers – this time, tech sales


This is still fairly trivial...

...not for those who are being 'victimised' though.

Back in the early '90s Lou Gerstner manged to remove about 40% of the workforce...something like 180 000 left 'Ive Been Moved'.

However it didn't really take that long to get the workforce back to 400 000. What rubbish leadership.

Calling your redundancy programme Baccarat? Immense Bummer, Management


Back when IBM was a good company

as a first line manager (technology) you typically had about 8 - 10 people reporting to you. Second level management had 5-6 reports and third level had 4-6.

So if there were 100 in the firing line you could at the very least expect to see some management take a hit at the same time.

pity that HR never seems to qualify for these 'career options"

Do the numbers, Einstein: AI is more than maths as some know it


Absolutely...the ability to subtract trip19, trip 17, trip15 and trip13 to derive the best way to get to double16 lives on and will never be forgotten. Bit useless for anything else mind you!

In 2012 China vowed 'OpenStack will smash the monopoly of western cloud providers!'


Huawei need to do something radical

Experiences with FusionSphere show that there is a huge amount of effort required to turn it into a viable solution for dev and test, forget production.

Huawei's ability to support FusionSphere is basically non-existent on our continent. (similar to their ability to support OceanStor etc).


Huawei need to do something radical

Experiences with FusionSphere show that there is a huge amount of effort required to turn it into a viable solution for dev and test, forget production.

Huawei's ability to support FusionSphere is basically non-existent on our continent. (simila to their ability to support OceanStor etc).

Oh, 3PAR. One moment you're gliding along. The next, you're in the rain as HPE woos Nimble


Re: Is it my imagination


thanks for sorting me out with my "D"yslexia. Always seem to mix up Data General and Digital Equipment Corporation, although both were very good and competitive companies when they existed.


Re: Is it my imagination


Thank you for pointing out some errors in my comments (or drivel if you prefer):

of course SVC was not based on Falconstor , it was based on DataCore SsnSymphony.

However the inband versus out of band comment is a bit off since Falconstor NSS runs inband (Ability to integrate with both FalconStor CDP for out of-band data protection and FalconStor NSS for in-band storage virtualization.)

The big issue I have is that although the IBM, HPE and EMC's of the world throw billions of dollars at research (not a lot of it on storage, storage software etc), they still go and buy their technology and with perhaps the exception of EMC fail dismally to incorporated whet they have bought (or OEM-ed) into a cohesive stratgey with all of the requisite ecostructures to support their clients adequately (reliance on partners!!!).

Dell EMC has gotten itself into something of a tangle with xtremio and DSSD, now Nutanix.

I did say the DS8000 was IBM through and through and I have only seen it in a mainframe environment, which unfortunately is a dwindling market place..


Is it my imagination

or has HPE lost the INVENT that used to characterise the old, old HP?

Certainly on the storage subsystem level it would seems so. Just about every product/solution they bring to market is not HP developed (wasn't the MSA a DG product??)

So, in my thinking, they are basically going the same way as IBM in losing any capability of developing leading technologies (I am convinced the SVC was based on FalconStor ). DS8000 is hardly an industry leading technology.

So it is confused your sales force with too many options, confuse your existing clients with too many options and no-one to unravel and develop a best fit solution (extremely poor pre-sales support).

regardless of what/ who they buy they are in trouble.

(Not a 3PAR fan either)

Dell EMC lifts the post-acquisition axe, swings


HR on the Chopping Block??

Never could happen as they are the Executioners and if the numbers to go are big enough they will need even more Executioners.

When it is over they all sit back, wash off the blood and pay themselves nice bonuses.

How I wish for the days of a Salary Department!!

What’s next after hyperconvergence?


Ah, the Mainframe...

Sure HCI systems are possible solutions for really stand-alone (silo-ed) applications. As soon as someone says I need that information from that silo-ed environment, the fun begins.

Mentioning mainframe brings back some very salient points regarding how they (at least IBM and compatibles) worked....partitions(I suppose modern speak containers) that could even communicate with each other, pass information, have differing priorities etc, let alone discuss mainframe virtualization technologies that still kick the crap out of any x86 offering.

Let's get Cloud out of the way as well, all it effectively is, is a Computer Bureau on steroids, with infinitely better network connectivity.

Enough of a rant...beer time and nostalgia.

No envy for NVMe: Hardened newbie talks to the Reg


Re: One small problem here

Absolutely, here we have a solution to a problem very few companies have. It doesn't seem much of a solution either.

Dell EMC World tease: What does 'composable' mean to you, readers?


Huawei today announced OpenSDS, an open-source project to replace vendor-specific storage controllers with the weight of enterprise and external storage number-one Dell EMC behind it.



SMI-S etc.

Ooh, Goody, more standards!

Boffins eschew silicon to build tiniest-ever transistor, just 1nm long


Re: "eschew silicon"

"S" for substrate, not an active component.

Scale scales: HCI maverick heads upstack with apps


“A rising tide lifts all boats.”

Had a colleague who was not particularly nautical. He moored the craft he had borrowed extremely securely to the jetty to ensure it did not slip away down into the estuary. Spring tide that evening saw the boat not rising with the tide and consequently it sank.

So Mr Collier's quote “A rising tide lifts all boats.” perhaps needs some correction to add in flexibility to react to unusual and changing circumstances (to avoid situations that NetApp seems to have got itself into..albeit not sinking so dramatically)

IBM is now selling Spectrum Virtualize free from hardware


If I remember correctly SVC started life as a rebranded FalconStor back in the early 200X years. They did some re-writing then bought Storwize and added that software to SVC, along with compression that came from somewhere else. Then they bought Texas to do the flash. Lot of IBM originality in Spectrum Virtualize!!

Beer merger dwarfs EMC/Dell


Several of SAB Millers european breweries have already been shipped off to Asahi from Japan

Thought the icon appropriate, but very concerned that the takeover will leave our SA beers alone.Heaven forbid that Bud comes to the African continent, even if they did sponsor the 2010 Football (not soccer) World Cup.

Microsoft joins the 1c/GB/month cloud storage caper


Re: Sounds good!

Now the bright blue cloud that Bill built has added a “cool” tier to the service that reaches the US$0.01/GB/month price once you store 100 terabytes in certain Azure regions.

Once you factor in the cost of the 100TB pre-requisite storage, then the arithmetic makes an even more ugly story.

Storage with the speed of memory? XPoint, XPoint, that's our plan


Re: That table again...

Seek time has nothing to do with rotational speed as it is a measurement of time taken to move the read/write heads from one track to another and average seek time is approximately the time taken to move over one third of the tracks on a drive.

Access time, on the other hand is a much more realistic and sensible metric in that it should encompass the mechanical and software related delays in the IO process and is applicable to ALL types of storage i.e . finding a tape in a tape librayry, waiting for an available drive and loading the tape and searching to the correct location for data on the tape......is a LOOOng access time whereas solid state devices are really quick.

Rotational speed is much more about data transfer rates, once the heads are over the required track, how quickly can it transfer data to or from the track.

Flash memory will send hyperconvergence to hyper-speed


plus ça change, plus c'est la même chose ‎!!!

It seems that every time a newer type of hardware technology emerges (or hits the marketeers) that the role of IT will never be the same again.

Over the last 50 plus years we must have heard it about every two years.....however the fundamentals of data processing (now ICT or whatever) have not really changed: There is/are workload/s that needs/need processing - no matter what type of workload it is. It needs compute, I/O , storage capacity , network and all sorts of ancillary services that have been developed over the years to deliver the end user requirements, effectively and efficiently, at an acceptable price (throw virtualisation in as a conversation piece---been around since the late '60s).

Solid State Storage is not new, we had customers with 16,24 and 32 MEGABYTE devices used for swap and paging in the late '80s and early '90s), very fast and VERY expensive but they did improve certain workloads that had high memory requirements.

The introduction of newer technologies that can have a significant impact on workload delivery is to be welcomed but until there is a complete revolution (extremely unlikely) in the way data processing operates it really is more of the same. Bottlenecks and contention will be with us forever, with the ultimate goal of making the actual workload the problem i.e. no more work to be done.

IBM slices heavy axe through staff in the US


Why doesn't Ginny Just Ask Watson for strategic advice?

Got Oracle? Got VMware? Going cloud? You could be stung for huge licensing fees



Must be x86 unless VMware now runs on SPARC.

OVM is the only hyperviser that Oracle will support for CPU affinity being set on.

Oracle is not so perverse on hardware partitioning but still be very careful.

Post-pub nosh neckfiller: The gargantuan Gatsby


Gatsby or AK47??

In Johannesburg these are known (in some areas) as AK47, Simple, Easy to Make, Reliable (you really have to try hard to cock it up) and DEADLY.


Gatsby or AK47??

In Johannesburg these are known (in some areas) as AK47, Simple, Easy to Make, Reliable (you really have to try hard to cock it up) and DEADLY. So you definitely need a few of these to mark your passing on.

All-flash array bake-off: Load DynamiX finds six AFAs go into four


Can anyone explain----

the detail in the table:

specifically what does K IOPS /Dollars mean? If it said K IOPS/Million Dollars then I would understand--

Vendor A (as an example) 144/5.9 = 24.4 rounded down = 24 and vendors B and C are 29 and 18 respectively. So how does Vendor D end up at 32?

Similarly Relative Throughput / K Dollar for Vendor D should be 4.88 not 3.8.

However if the cost price is not 1 000 000 but rather 1 280 000 then the numbers make sense.

A further comment is that if the SLA requirements are specified clearly and require 100 K IOPS at a latency of not greater than , say 2 milliseconds, then the choice comes down to A or marginally B.

Or just change the SLA requirement to suit the cheapest solution anyway

BOFH: How long does it take to complete Friday's lager-related tasks?


Re: admin obverhead [sic]

Back in the '70s IBM EMEA (in its infinite wisdom) decided that it needed to understand , in detail, what the Systems Engineers actually did ----- even though any customer could have told them. So, out came the work item and time allocated sheet---work items were coded to ensure compliance across the board (and the BORED).

One of the codes was 'Forced Idle Time', which was meant to specify time sitting around waiting for the last job to finish before you could start your standalone time at 02:00. Just about every engineer I knew used that code to great effect when filling out the form....seemed to take 2 days and fortunately it could be filled in while offsite (read lager time) and no-one was the wiser.

This lasted something like 3 months before it was canned.

Mainframe big boy Big Blue tries to drum up new biz via Linux


Re: Emperor or Rockhopper, nothing in between

I think your are on the wrong track here.

As mentioned in an earlier comment, Emperor and Rockhopper are penguins, so give the IBM marketing guys some credit for linking the names of the systems to the Linux mascot.

If you are looking for an intermediate sized MF then perhaps the 'Jackass' penguin could be a suitable name!

If you are Italian then the 'Macaroni' might suit you.

Huawei and DataCore spawn a beautiful hyper-converged system


I agree that this combo is not really hyper-convergence but I am sure that this is more about Huawei than DataCore. At the moment Huawei does not appear to have a viable go-to-market model for Converged Infrastructure (note not hyper-converged), and using DataCore certainly addresses some of the major storage issues they face. It adds a different dimension in that theDataCore software opens up a whole slew of opportunities for storage convergence, especially for legacy arrays (I hate that expression).

SpaceX blasts a mischief of mice, a 3D printer and a cuddly toy* into SPAAAACE


MICE vs. Geckos

The Russian experiment with sex-crazed Geckos failed frozenly!

Do the Americans think they can outsex the Russians using warm-blooded horny little buggers called Mickey, Minnie et al?

Heaven help the Space Station if they get loose - or are they all infertile or the same sex?

LOHAN acquires aircraft arboreal avoidance algorithm acronyms


Re: It's not just LOHAN

I am sure that something from Peanuts is applicable here:


Boffins: Antarctic glacier in irreversible decline, will raise sea levels by 1cm


I cannot believe that no-one has mentioned the Ice Ages of which we have had plenty (and well recorded through enough scientifically 'provable' evidence) and maybe climate change is actually how the Earth has managed to get itself to where it can support 7+ billion people (who have no regard for it or the other life forms extant!! )


Re: A choice of words

Not necessarily the best comment ever, but it does highlight the real requirement for questioning and confrontation.

Boffins claim battery BREAKTHROUGH – with rhubarb-like molecule


Re: Optimistic +1

Rhubarb wine IS nice....not supposed to be.

Just takes about 5 years to ferment/mature.