Consolidated Business & Enterprise Computing Rant Thread

Discussion in 'Business & Enterprise Computing' started by elvis, Jul 1, 2008.

Tags:
  1. DavidRa

    DavidRa Member

    Joined:
    Jun 8, 2002
    Messages:
    2,911
    Location:
    NSW Central Coast
    3x? Don't know where you're getting them that cheap. In my experience it's nearly 10x on a lot of that stuff.

    Complete joke.
     
  2. Daemon

    Daemon Member

    Joined:
    Jun 27, 2001
    Messages:
    5,144
    Location:
    qld.au
    On one hand, I want to agree and applaud.

    On the other hand, it's 110% accurate and therefore not a rant.... I'm torn.

    Currently seeing pricing on 50+TB of SSD storage and getting $vendors trying to argue their kit is worth a 50% premium over the RRP of the underlying drive.... not including the chassis. Not even a single yacht invite either :(
     
  3. bcann

    bcann Member

    Joined:
    Feb 26, 2006
    Messages:
    4,465
    Location:
    NSW
    Coulda swore i read somewhere the SAN dropped a whole disk shelf.
     
  4. Perko

    Perko Member

    Joined:
    Aug 12, 2011
    Messages:
    1,852
    Location:
    NW Tasmania
    Yeah I thought it was far more than one disk too, but the spin, (no pun), has started now. Unless they can prove that it's a physical or firmware fault on the SSD units themselves though, they're just pissing in the wind doing this.

    But I guess that there are enough non-technical, purse string holding decision makers in the world to throw a bit of plausible deniability into the mix. After all, the E stands for Enterprise, so it must be super-special business grade.
     
  5. OP
    OP
    elvis

    elvis Old school old fool

    Joined:
    Jun 27, 2001
    Messages:
    30,852
    Location:
    Brisbane
    Assuming the DBA/dev didn't fuck the indexing up, and your database supports full text indexing, or you don't want to do something crazy like regex patterns... pretty much if all the stars align and the tools let you do the trivial thing you want to do, then it works.

    Yeah, it was a nice try. Still "this specific thing for this specific need" territory, rather than "just make it fucking easy for people to do whatever they want", which I prefer (because my wants aren't necessarily your wants, nor the wants of the developer).
     
  6. wintermute000

    wintermute000 Member

    Joined:
    Jan 23, 2011
    Messages:
    1,056
    Well there IS the thing re: enterprise providers guaranteeing that you can still get a copy of component XYZ that is (in theory) exactly the same as the one in your tin up to say 5 years from now, whereas if you were sourcing components directly they'd probably be out of production quicker than that.

    E.g. an Enterprise rebadge of say a Sammy Evo 840 (I know I know not enterprise not SLC or decent MLC yada yada its just an example) would have the firmware and flash and all that jazz locked in for 5 years. You try getting a new Evo840 now, let alone in another year or two's time.

    Now having said all that I'm not in the hardware vending game so I have NO idea how much cost this would add, but I presume it would not be trivial?

    Also the testing/validation of component XYZ against your custom hardware. Sure a lot of the time, esp. interfacing through standard APIs/protocols it should just work, but the THEORY (again IRL standards, er, vary) they should catch all the exceptions and issues and give you nice 100% compatible no surprises behaviour. Even if 95% of the time the whitebox part works perfectly, that 5% of the time it blows up in an unexpected fashion could end up costing you a lot more money than the initial saving.
     
  7. Perko

    Perko Member

    Joined:
    Aug 12, 2011
    Messages:
    1,852
    Location:
    NW Tasmania
    Sure, but is that worth, (using the anecdotes here), 300-1000% of the coin? Particularly given what's happened with the "genuine" kit. White box supply chains still have the wholesaler/retailer to go through, so you could probably assume that the enterprise vendors would achieve a decent economy of scale in terms of price per unit by buying direct in very large bulk numbers, even if they had to buy warranty units at a rate of 1:1, (which is unlikely at best). This case is a screw up that reflects badly on the segment, but the underlying premium STILL results in smaller, yet data reliant businesses forcing dodgy Bob down the road to plug a few extra WD Blues into a PC that they bought from Harvey Norman, RAID them up, and call it a storage server. This is so bad for the long term reputation of IT in business, because name me a business person that would blame themselves and tell all their associates that they made a mistake by going cheap, rather than "f####ing computers, and f##k that dodgy Bob guy too."

    I see the same thing in agriculture, where vendors of decent kit are actually holding the industry back by demanding extreme money for it, driving the small to medium guys into the arms of bargain basement imports that introduce serious safety risks to the operators and people around them in some cases, as well as resulting in a very slow trickle down on the introduction of productivity gains through precision ag and other techs.
     
  8. tin

    tin Member

    Joined:
    Jul 31, 2001
    Messages:
    6,402
    Location:
    Narrabri NSW
    The bean counters overruling local decisions in NSW schools back some years used to use a similar argument for the printers we were forced to buy.... Double to triple the normal wholesale price, but double the warranty...
    Never occurred to them that you could buy 2 regular price printers, and hold one for your self. Or shock - it's a printer - buy a new one if the old one dies just outside the default warranty since you just saved the cost of a printer anyway.

    Point being, if you're paying double or triple, you can buy double or triple the products and store them for if something goes wrong. Then YOU are assuring the parts are available. I've seen a certain 4 letter vendor starting with "A" send completely different parts for "enterprise" warranty replacements, so trusting the vendor isn't that good of an idea either.
     
  9. GumbyNoTalent

    GumbyNoTalent Member

    Joined:
    Jan 8, 2003
    Messages:
    4,791
    Location:
    Briz Vegas
    Holding stock costs money, retrieving the stock costs money, having the expertise (thin excuse) to use the spares properly costs money.

    You have also forgot that "enterprise" level hardware agreements usually mean they will come to you and replace broken part within 4,8 or 24 hrs depending on how much you have paid.

    IBM still sell service agreements for hardware they sold 20+ years ago, sure those agreements costs more then the price to replace the hardware but if the customer wants to keep a 20+ year old RISC6000 running they can (this example was based on a real world example I stumbled accross), and IBM will facilitate them just like HP and Dell would their own hardware.

    Bean Counters prefer the risk be with external party, thus the whole outsourcing movement of the past 10 years, same applies for hardware, as the company usually bundles the entire cost up including support and makes it a service contract including the hardware.Printers are a classic example you can now contract the entire service printer, toner, paper and servicing to a fixed cost, a bean counters wet dream.
     
  10. millsy_c

    millsy_c Member

    Joined:
    Mar 31, 2007
    Messages:
    11,682
    Location:
    Brisbane
    You can't outsource risk ;)
    People certainly try to but yeah you can't formally do that
     
  11. GumbyNoTalent

    GumbyNoTalent Member

    Joined:
    Jan 8, 2003
    Messages:
    4,791
    Location:
    Briz Vegas
    Management 101 : Move the risk/blame for bad things happening to an external source, thus blame shifted else where.
     
  12. millsy_c

    millsy_c Member

    Joined:
    Mar 31, 2007
    Messages:
    11,682
    Location:
    Brisbane
    Which is patently incorrect, you can shift the blame but you still have to be accepting the risk of it. Any old management consultancy or auditor will lump bulk shit on you for doing that
     
  13. cvidler

    cvidler Member

    Joined:
    Jun 29, 2001
    Messages:
    10,862
    Location:
    Canberra
    exactly.

    You're customers trying to buy your widgets/services don't know/care it's HP's fault, your systems are down, and they'll go elsewhere.

    Govco is lucky they're a monopoly, no one else can do your BAS/PAYG etc.
     
  14. GumbyNoTalent

    GumbyNoTalent Member

    Joined:
    Jan 8, 2003
    Messages:
    4,791
    Location:
    Briz Vegas
    Mine is an observation, if you have observed differently then I wonder why so many posts in this tread convey the same observation?
     
  15. millsy_c

    millsy_c Member

    Joined:
    Mar 31, 2007
    Messages:
    11,682
    Location:
    Brisbane
    I am not saying that plenty of managers *think* they can do that, I'm saying from a risk management perspective it's just flat out wrong.

    E.g. IT goes to the cloud to reduce the likelihood of encountering an outage. Just because the cloud vendor now maintains the systems doesn't mean the risk changed, just the likelihood etc has changed. The IT manager still "owns" the risk though
     
    Last edited: Feb 19, 2017
  16. wintermute000

    wintermute000 Member

    Joined:
    Jan 23, 2011
    Messages:
    1,056
    You're both correct.

    What Gumby says reflects the attitude for tonnes of customers.

    Must be great to get paid the big bucks to decide to go to AWS/Azure. Geeze must have taken a lot of brain cells to work that one out.

    I cringe every time I have to sit through some stuffed suit drone at me for 20 mintues re: their 'cloud strategy' when its basically yes you are a typical hostage to MS enterprise and you will just use all of MS's azure stuff, wow how surprising, what's your pay packet again for 'architecting' all this (and no doubt taking all the credit for a 'cloud project'). They need to be locked in a shipping container with all the ITIL and TOGAF people, towed out to the Mariana trench and simply dropped into the ocean

    Disclaimer: I have nothing against Azure or AWS, some of their xaaS stuff is really super neat and I heart our DevOps overlords

    back on topic, I understand the 'owning the risk' part - its no different from classic Westminster system ministerial responsibility - but IRL what can you really do if you've gone for cloud, as long as you do the sane thing and provision in different availability zones etc. and when going dual provider will blow out the cost and complexity by huge amounts (different API/tooling, pay for moar services). In a world where you can't even get the bean counters to recognise the 'good, fast, cheap, pick 2' principle. Remember that big Sydney AWS outage?

    Basically what I'm asking is: to reduce the risk beyond single provider multiple zones blows the cost out. Yet when an outage inevitably happens they will always come running back and asking 'why isn't it MOAR redundant, no we don't remember complaining about money EVAR, what is that requirement spec I never signed it, oh you mean REQUIREMENTS'
     
    Last edited: Feb 19, 2017
  17. GreyWolfe01

    GreyWolfe01 Member

    Joined:
    Aug 1, 2001
    Messages:
    2,198
    Location:
    Sydney, Australia
    Good to see that my pet cloud hate is alive and well in the community.
     
  18. OP
    OP
    elvis

    elvis Old school old fool

    Joined:
    Jun 27, 2001
    Messages:
    30,852
    Location:
    Brisbane
    Dude, I love the cloud.

    Every month there's a bill for exactly what we use, itemised by department. Finance piss and moan, and we remind them that it's not IT, but the business who are using this stuff, and the true cost of computing is revealed.

    Finance pressures departments to cost-cut, so they seek alternative cloud vendors. Migrations are costly and difficult, often causing massive productivity outages. Staff are upset, everyone hates the cloud pushers.

    Meanwhile, IT remain totally unscathed. The business is forced for the first time in forever to actually consider the ramifications of their fickle, wasteful nature, and their inability to plan more than a week in advance is exposed to everyone.

    Dude, I fucking love the cloud.
     
  19. NSanity

    NSanity Member

    Joined:
    Mar 11, 2002
    Messages:
    16,503
    Location:
    Canberra
    FFS, you don't even need to do active multi-zones - just backing the damned thing up to another provider will trip past the $.
     
  20. GreyWolfe01

    GreyWolfe01 Member

    Joined:
    Aug 1, 2001
    Messages:
    2,198
    Location:
    Sydney, Australia
    By that you mean that IT is blamed for any problem, even though now the risk and responsibility is with an outside vendor?

    IT is never unscathed. It's always our fault.
     

Share This Page