Is the Sun shining?

Today Brian posted about Sun and their Sun Rays and thanks to that and to some comments from crazy people (yes, me), it caused quite a reaction within the Sun community.

First of all, it was simply a post, not a review of anything. And my comment there was specifically tailored to Craig Bender, a guy that I had great conversations in the past and for some reason (not sure if I pissed him off, if I said something not to be said, etc) he simply vanished. I knew if I mentioned anything like ‘SunRays suck’ or ‘SunRays are the reason why Sun had net losses of US$ 1.677 Billion in the first quarter of 2009′, I would be able to check if he was still alive. So minutes after my comment, BINGO. The fish got the bait. 🙂

So history and jokes aside, here is the deal. I know some people think that heterogeneous environments are the norm out there these days but in a way I disagree. At least based on my work on the field doing consulting services, most of my clients are indeed a Windows shop. Some do still have Novell (surprised?) but in most cases getting rid of it. And some do have some Unix but usually only for very specific back end requirements.

This brings us to this: for the user, even though there may be Novell, Solaris, HP-UX, etc behind the curtains, he has a Windows desktop and/or accesses one hosted either on TS/Citrix or as a VDI offering. His apps, even if they access data on a database running on Solaris, still has a Windows or web front end (running on a browser on top of Windows). So for all the user cares, this is a Windows environment. Does not matter you tell him all the databases are on Solaris running Oracle X. He does not give a crap and does not know it. I guess it is from this perspective that Brian made the comment he was a Windows guy.

For sure on the backend it is another story. It can be a mixed environment and probably is, considering Linux is free and I think even Solaris is free now. But even in this case what I see is usually the ‘Unix guys’ are not the same ones managing ‘Windows’. Same for the ‘Oracle’ people and so on. So still, you may be indeed a ‘Windows’ guy in a place where all the databases are hosted on Solaris servers running Oracle. You simply do not touch/see them at all, even as an administrator on the ‘Windows’ side.

Back to the title, after seeing all the losses Sun posted, no matter what they say they were probably not doing good. My accountant tells me if you are doing great you usually do not post a US$ 2.234 Billion net loss for the year with total net revenues reduced by US$ 2.4 Billion (compared to 2008 revenues).

Now with the Oracle acquisition, what can Sun really do? I have always seen their Sun Ray products as a big niche but there are no numbers out there to compare them to other thin client vendors (again, I am just trying to compare units moved and not really if they are better than the competition for reasons A, B or C). May be Craig will jump in and provide us with some numbers for comparison so I can prove (or proved wrong) they are a niche, on a niche market.

The bigger question I have is, is there anything Sun can really do to make an impact on this market (SBC/VDI) or is their fate to always be ‘another minor player’, ‘niche solution’ and so on forever? Maybe Craig is correct in all they may need is good marketing. Their Appliance Link Protocol is indeed cool and lightweight but is that enough to make SunRays widespread? So far, in 10 years since their introduction that is not what I could see. But again, if it has been a commercial success or failure it depends who you ask. Not sure how much money they invested over the years to get to what they have today and what was their revenue with such product line since 1999. This kind of number they usually do not post…

Technically I would love to spend a lot of time with their stuff, if Craig is willing to help me on that end. In exchange I would write an ‘Honest Opinions’ article, like my presentations at BriForum. An unbiased, no bullshit review of whatever I see, pointing out the strengths and the weaknesses. Or even present a session about my findings on the next BriForum! Why not?

If anyone has links to the products I will need to download to try their stuff, please go ahead and post them. For the hardware, again, I will need help from Sun to make it happen, assuming they want to make it happen.

And Sun people, my apologies if I pissed you off with my comment on Brian’s website. It was just really a joke to grab Craig’s attention.


1,787 total views, 1 views today

2 Replies to “Is the Sun shining?”

  1. Taking the overall loss of Sun as an indicator for the fittedness of a specific product for a given problem, is simple bullshit… 😉 And is only trying to discredit something for something else… What’s it, you are jealous of in Sun’s VDI solution?

    OK, with that of my mind ( 😉 ), let’s state:

    Sadly, Sun’s annual report does not specify numbers neither for VDI nor Desktops nor Sun Rays.

    But, now, back to the question at hand:

    There’s more to be checked, if comparing VDI solutions, then only the underlying OS for a small, although important part of the overall solution/offering.

    Some of these additional topics include:

    1.) Costs of acquisition (what does ist cost to build an environment, HW and SW and installtime)
    2.) Operating expenses (and yes, there you have costs for admins, if you need a “new” OS platform)
    3.) Security (overall, starting from separation of users to separation of processes to security against intruders on the overall chain of devices and software stacks)
    4.) Efficiency (where do I get most power for the buck)
    5.) Access point in case of problem (one point shopping, one point service?)

    I’m sure you read about: <a href=””< Sadly, that did not include Sun’s VDI. Still, there is something to learn here. The differences between the different solutions (TS, or Xen, or, …) are not the decision making points, as the differences are not as big. A main finding there is, that memory per server is a limiting factor. That’s why for example, in former times, the Sun X4600 M2 was a very attractive system for large VDI environments. And, as further answer to 4.) above, Sun’s VDI allows you to run either VBox or VMware as a basis for user sessions, which also influences points 1.) and 2.)

    For point 5.) only MS and Sun can offer a “two stop shopping” VDI solution, MS by adding a single HW supplier, and Sun by adding MS licenses. All other VDI vendors need a three stop offer, as VMware does not own MS licenses nor Hardware, and as Citrix also does not own HW nor MS licenses. That is a point, that’s not to be underestimated! And here Sun’s the only one how can offer two stop shopping for HETEROGENEOUS environments.

    Now, let’s look at point 3.), security.

    We all know, that Solaris is the most advanced and efficient (scaling nearly linearly with addition of CPUs way beyond 100 CPUs) OS on the planet. This helps immensily, when defining large scale environments, because consolidating onto large systems gets possible, because the OS is not simply managing itself and the underlying resources, but leaves many space for the apps. With putting for example every single VBox instance into a separate Solaris 10 Zone/Container, you additinally get the benefit of fine grained resource control AND security, as that environment simply is not able to break into a different Zone/Container. And Zones/Container are EAL4+ certified… 😉 (afaik).

    And, an additional topic for 4.) is putting the VDI images onto ZFS. Cloning get’s easy and quick… But that’s another topic…

    So, I would love to see projectvrc results for Sun’s VDI… And a more vivid discussion about the pros and cons of VDI solutions in total…


    1. Hi Matthias!

      As you could read, I never said on my post that every single division at Sun is not making money. Overall, as a whole, they were on the red and had, as their numbers clearly show, a very tough year. When a company is losing such huge amounts of money, usually all business units get affected, even the ones turning up some profit. Note the word ‘usually’ as that may not be the case here. And one more time, I never implied on my post all divisions were in the red.

      Back to the topic of the post, even with your comment, the main question remains: with all the good stuff you pointed out, can they still make a difference on such market and actually become a major player on it or are they fated to be a niche player? Keep in mind that not always the best technology or most advanced product wins the battle. The computer industry is a good example of that. There are far superior OSs than Windows and all their variations (like Solaris as you pointed out) but the number of servers sold in the world that will end up running Windows or Linux is far superior, by very large numbers, than Solaris. We like it or not, that is the reality.

      I do not have numbers for 2009/2008 but for example, IDC data shows that worldwide Linux shipments in 2006 were about 2.4 million and nearly 2.7 million in 2007. By contrast, Solaris shipments totaled 376,000 in 2006 and 371,000 in 2007. Even compared to Linux, these are indeed very small numbers and therefore my notion of ‘niche’ I am certain still applies (just imagine out of these 371,000 shipped in 2007, how many ended up being Sun Ray servers? Probably a very small percentage).

      So again, considering their past performance, numbers and marketing strategies so far, can Sun become a major player in this space? Time will tell.

Leave a Reply

Your email address will not be published. Required fields are marked *