Silicon Valley has a new parlour game: what's going to happen to Sun after the cuts? Rumours suggest it will merge with EMC, that HP or IBM will take it over, that Fujitsu will buy its hardware business, or even that the StorageTek storage business unit will be spun off. Do Sun's leaders want to be redeemed or to retire? The …
Impairment of good will
1.65 billion of Sun's loss was a writedown due to new and fairly complicated accounting rules, called a "goodwill impairment'. Apparently, the intention of the rule is to make CFOs think twice before the company undertakes a merger or acquisition.
If you look up goodwill impairment, it's kind of obscure in explaining how it's an actual loss. It looks like it's really just a loss on paper.
Without that, the loss was 65 mil.
Also, when I looked at the quarterly report, I'd seen that the cash in bank was 3.1 bil., not 2.3. Am I missing something?
Youre making a huge assumtion...
Why does Sun need to be taken over or sell a part of itself? There is simply no reason. It's doing fine with $3bn cash in the bank - sure the high-end is taking a dive right now, but the newer products are ramping well. It's just a transition thing and needs about 12 more months to show the true potential.
You say Sun lost $1.7bn the past quarter but most of that was an accounting adjustment for a goodwill write-off. The "true" loss was $60m on revenues of $2990m (i.e. 2% of revenues). It's just a matter of right-sizing the SG&A so that revenues always come in above expenses. Sun might need to scale back its $2bn yearly R&D to something more like $1.5bn or less. That's still a huge R&D budget!
Sun needs to ignore "in house" software, and concentrate on hardware to survive.
Sun has tried to be a "network in a box" company for over a couple of decades, but in my experience it's only been the hardware that was worth buying ... I personally have Sun kit from 1987 that hasn't missed a beat in 20 years, but it's all running one of the BSDs ...
"then it's more likely they'll sell off a lump of Sun that is not central to that strategy." With EMC's Toxic Appraisal of the Infinite Open Source Field, there would appear to be a MisMatch .... Producing a Blockage/Blockade/Conflict.
What is the Sun Driver, .... a Long Shot with the EnigmatIQ Algorithm? 42Entertainment Apache Territory? :-)
Explain term "proprietary"
This term, used without explanation, annoys me a little, I mean, almost the whole Sun software stack is open source, including Java, Solaris, the MySQL database and the vast majority of it's middle ware. Add to that the fact that their Niagara processor line is also open source (you can download the Verilog designs and fab your own if you want). Compare this to a typical Microsoft/Intel combination, none of which is open source or even based on any kind of ISO/open standard, but yet the word "proprietary" would never be used to describe it.
Personally I care not whether software is FOSS or not, but I'd like to see justification for describing Sun kit like this, when in fact the Solaris/SPARC/Java/MySQL combination is probably the most open and freely available system you can buy.
How does having a cash pile equal to market cap make the company worthless. If IBM has a market cap of $112ish billion (as you stated) and cash reserves less than $112ish billion does that make IBM worth a negative amount?
I see why you're saying it, it can just about cover the money shareholders are "owed" but are you seriously suggesting that any company with cash reserves less than market cap is a bad bet? I'm sure after Google's IPO it didn't becomes worth less and less as the share price rose, did it?
RE: Youre making a huge assumtion...
I have to agree with Kevin here. The whole market is in the pits at the moment and few tech companies are going to avoid the current downturn. You only have to look at the share price of IBM, HP, MSFT and others to see this and Sun is no different. Sun unfortunately was still in the process of trying to recover its stock price when this downturn hit and with all the short selling of stock that happens these days (should be illegal really) it really give a false impression of how a company is performing. You only have to look at Sun's revenues to see how much cash it brings in each quarter, the issue is to reduce costs and improve profits which it is doing. Sun has some awesome products right now and it is set to come out of the downturn on top if it gets things right.
As for someone buying Sun, its possible but I don't think anyone can afford to splash out cash right now during these hard times. A takeover would be bad for competition and ultimately for the customer and the innovations that Sun comes up with.
Well, it is not only a economical problem
Well IMHO their problems are more fundamental than many think, I personally find that SUN is not looking after their customer base. Which is very loyal, some to the point of fanatics. It seems to me that they have raised prices on the boxes that they normally sell to their loyal customer base. And this will surely generate more revenue, for each box they sell, but it will also scare some away.
IMHO SUN's loyal base is the lowend SPARC servers that run Solaris. There are A LOT of those out there, and you used to be able to get a decent performing box, to a price that was lower than it would cost you to get one from HP or IBM.
Sure the HP/IBM box would be faster. But the difference in price in the favour of SUN and the fact that the IBM/HP server wasn't that much faster, meant that there really wasn't any reasons to migrate to something else.
That have IMHO changed.
Now if you today have a v215 or V125 or even an oldish v4X0 you want to replace with another SPARC server, then You basically have 2 different choices Buy a Niagara based server/Blade or buy a M3000.
The same kind of money that gave you a v215 will now give you a a 4core 1.2GHz T5120
And the money that gave you a v125 will now get you a 4 core 1.0Ghz T1000.
This is not an ideal fit for many SUN customers, as you with the Niagara based servers might end up having worse single threaded performance than on your old server. And for many workloads that will mean worse throughput.
That leaves the M3000 as the other option.
Now the M3000 is 15K USD and it's not that fast. SUN just released a SAP2Tier benchmark of 4130 SAPS. That is a good deal less than old submissions from HP and IBM.
The rx2660 using Montecito and the p505Q using Power5+ both does around 5500 SAPS.
Now the rx2660 would be a good deal cheaper, if you choose to use lower clocked Montvales rather than the highest clocked ones, and would most likely cost around 10K and a p505Q is 5.5KUSD.
This means that low end servers from HP and IBM now aren't just faster but also cheaper than their SUN counterparts.
This is bad for SUN, and we haven't even started to talk about Linux taking marked share yet.
And what about their server product lines ?
There is a install base out there with SUN Enterprise and SUN Fire servers who have been waiting for a replacement, in the form of Rock based servers. And if you've ever talked to SUN sales people then many told their customers to jump over APL servers and go directly to Rock based servers.
And well now it looks like SUN will try to make these people move to MX000 systems:
And I would imagine that this would mean that HP and IBM will unleash the sales dogs of war, on the customers that have oldish SUN equipment, and either take business away or force SUN to discount more than it would have liked to do.
So what will happen to SUN ? Hopefully the will survive in some form as the IT industry will be a poorer business without them. But they need to change things. I agree with the with Cris Mellor (the writer of the article) that I don't think that neither HP or IBM would pick up SUN. But a alliance between Fujitsu and IBM might, where Fujitsu took over the Hardware and IBM some of the software division, but I don't know how much money IBM would actually pay for getting their hands on Java.
Sun Software will survive (new name as they drop Microsystems)
Let me answer a few questions then I will give you the scoop.
Goodwill? - Yes it is an GAAP requirement to write down bad acquisitions when your own value is less. You could say its only accounting but it is a public statement that Sun admits its huge $4B acquisition of STK was a complete waste of cash. You have to be accountable to your stockholders. Pony decided to wipe out all STK goodwill vs. being upfront about each acquisitions current value. MySQL is now priced about $300 but still carries the $1B of goodwill.
Kevin. Sun is obviously not doing well. Any company who spends $3B in stock buybacks from its cash and the stock value decreases 85% in less than a year is not doing well. Shareholders are leaving and they are the owners of the company. True loss of $60M in in quarter may not seem much but Sun cannot make money according to its track record. Personally I think they actually lost $3B + $60M since they wasted that cash on stock buybacks and it did nothing for the value of the stock. As far as the $2B of R&D it has always been a statistic of pride for Sun, but keep in mind Sun only had 600 patents last year thats $3.3M per patent. Obviously there is rampant waste in the r&d department.
Jake. Wow you suggesting the exact opposite strategy that Sun needs. Sun's hardware business is why they are in such dire straights. Their only good products are the low end niche T boxes. They are not considered a player in the x86 space as their only positive attribute is cramming components into half the space which does effect reliability. Their enterprise systems are nothing more than OEM Fujitsu gear. The right strategy is software for profit, but Sun does not have good products and opensource does not fix poor software products.unless you have a substantial community like linux.
So what is the future? Sun will survive because Pony has no intention of falling on his sword. Scooter still views Sun as his baby and will make sure something is left which is called Sun something. Pony knows the only way he can survive is if he breaks up the company. He reorganized the company to be able to sell the parts.
Pony's strategy is this:
>SPARC/Solaris ($1.2B) go to Fujitsu, IBM or HP so they can fight over the hardware install base. Fujitsu just spun off their money losing chip business and they have never been interested in buying an American company. That leaves IBM or HP. Why would Sun give up its best asset Solaris? Because no company would just buy the SPARC business and be held hostage to the licensor of Solaris.
>x86 will go to Dell. ($300M) Dell has poor enterprise technology and would benefit from getting into the profitable space and providing technology innnovation
>Storage/Tape (STK) ($850M) This will go to EMC. EMC does not have a tape product and could make money on some of the technologies Sun struggles with selling.
Sun will go private with the help of KKR and Southeastern and survive as the opensource software company with key products: JAVA, MySQL, xVM, OpenOffice company.
(And Pony will stop doing his 5minute Friday pep calls)
Not the time for takeovers
Nobody will buy Sun in the near term. Nobody will be able realise the liquid assets to do so until the financial system sorts itself out. That is why the market valuation is so low: not only is there no chance of a takeover, the cash mountain is counted below face value (that is, it exists but is not liquid).
By the same token, the cash isn't liquid enough for the company to buy its shares and go private. A higher share price would be nice, but the same lack of credit which is depressing its share price is also stopping any potential predators, which in turn is depressing the share price further. In the meantime, after a pretty bad quarter, the operating loss was $65 million against a cash mountain of over $2bn.
There's nothing to see here: unlike retailers, Sun has a large revenue stream from its large installed customer base, so sales won't fall over a cliff (though they are down). These customers have staff who are skilled in Solaris, not AIX or HP-UX: IBM or HP can't get into these accounts just by offering discounts.
That said, I don't for a moment think that HP or IBM would want to buy Sun's customer base, they want to take it without paying for it.
Hear, hear. We just buy Sun for the kickass hardware. Why would I buy an '80s era OS when I can get *BSD or Linux for free? SunOS doesn't even come with a compiler any more! A development environment is an add-on. On a UNIX.
Plus Sun (McNealy especially) has never believed in Open Source and it's just a marketing gimmick to them. "Oh well, everyone else is doing this open source stuff, I guess I can put on a penguin suit or something". You can see their true attitude toward Open Source with their Java licensing. As far as I'm concerned they're just another proprietary vendor wanting lock-in.
It's not obvious (@jake) - Sun's software is outstanding
You're so dead wrong. The hardware is good, that's true, but it's in software that Sun is not only leading the way but in fact blasting away all opposition.
The Solaris 10 operating system has been a revelation - as well as the high-profile items like ZFS and dtrace, it introduced zones, easy to use, very powerful and near-zero overhead virtual machines with resource management (Containers); the Service Management Facility, a complete rewrite of and vast enhancement to the /etc/rc.d|init.d system that have bogged *nix systems with shell scripts for years; the Fault Management daemon for error reporting and fault fixing across the system; and much more. Solaris 10 has enough significant enhancements to fill two or three major releases on most other vendor's schedules.
ZFS, DTrace and Zones/Containers in particular represent an order of magnitude over what went before, and over what any rival vendor is developing. The Linux kernel team and their numerous distrib cohorts haven't introduced an innovative, concept-changing feature in years (did they ever, besides price and being open source?), AIX is rock solid and very manageable but lacking any real innovation, BSD is reliable but under-resourced, HPUX is stagnated, etc, etc.
The world needs Sun software, they are the only truly innovative operating system vendor out there. I sincerely hope that they don't sell, or that if they do they find someone willing to be hands-off on their software R&D and let them continue on their path.
Without them, we'll be back to the mindless mediocre feature tweaking of every other OS producer out there.
Sun, just keep things simple
Sun keeps making the same mistake over and over again. They overcomplicate things and their market fails to 'get' it. I talk to customers day in, day out, and they are genuinely shocked to discover that Sun have 1-2 socket Windows-compatible servers. "I thought they did very bespoke high-end servers, I had no idea they played in this space" is the reaction that I see time and time again. And they have played consistently in this space since the V20z around 4 years ago.
And so begins a conversation explaining all the cool and usable things Sun have done in the last couple of years. It ends with the customer feeling very excited that suddenly Sun is once again a viable vendor for their next project - everything from server refreshes to identity management implementations.
Another point - I've seen dozens of adverts for IBM kit and products on television (in the UK) over the years. I've seen precisely zero adverts by Sun. For some bizarre reason Sun don't see any benefit in advertising on television to a UK market. Big mistake in my opinion. They need to a) break down that perception that they only do high-end specialist stuff and b) they need to spread that message far and wide.
Until then they will keep on bringing great new ideas to market, and the market will mostly continue to fail to grasp them. Lets's see how many people have heard of the open storage products in a year from now. My bet is they will continue to refer to it as Amber Road and not advertise to the masses - I can almost see the customers' eyes glazing over as I type.
RE: Kevin Hutchinson
Because of simple economics - Sun needs money to invest in product development and just to pay the running bills, including ongoing legal fees. Sun currently has four sources for such cash - product and services sales (declining overall, so more money going out than coming in), loans from the banks (and the banks aren't lending), sales of shares (nobody wants them, even at the low prices they are at), and by using the cash reserves. Sun has been burning through the latter just about continually since Y2K, and simply can't afford to continue doing so, not if it wants to pursue the patent troll route as it is doing with NetApp.
You want Sun to buy itself out? So, Sun uses it's cash to try and buy back shares, but what if Sun ends up unable to buy all the shares? It is still on the stock market, and now with no cash reserves. Which would make it even more of a basket-case, and the vultures move in for the asset-stripping in shorter time.
Or, Sun can wait it out, and hope the declining sales and cash reserves manage to last through the recession until some mythical boom time when customers aren't going to give a hoot about costs and just buy Sun kit for fun - unlikely, cue the vultures a bit later down the line.
Or Sun sells some assets to cut costs, maintains the current product lines but slims down on expenses, and hopes for that same mythical boom time. Only problem is they have very little assets of value left, even their patent portfolio would not raise a billion in cash, so they just end up opening the door to the vultures in stages.
Which brings us to what the market are really telling Sun to do - downsize and radically re-align.
And that means taking a long, honest look at the enterprise server business. Maybe keep the enterprise SPARC line for now whilst the majority of the development costs are footed by Fujitsu, but drop Rock and concentrate on x86 and possibly Niagara. If Niagara can't make the profits required to at least fund itself and future developments, be smart and drop it now. Talk to Fujitsu and find out what is really planned for SPARC64, and then make a grown-up decision about possibly porting to another architecture such as Power or Itanium (the former is now quite possible with IBM buying Transitive) if they want to stay in the enterprise high end.
If StorageTek can't make any money then take an axe to it and clear out the deadwood products. I have no idea why Chris thinks anyone would buy the tape bizz, especially HP (tape market leader)! Again, the patent portfolio is worth some but not much, but patents cost peanuts to maintain and can be easily sold later without including painful staff cuts, unlike hardware product lines. Tape is dying - apply investments accordingly, and treat as deadwood if required with calm hubris rather than the usual Sun posturing.
And Solaris - supposedly a crown jewel but one that can't seem to make Sun any money. Sun will just have to evaluate whether it can make money going head to head with Red Hat and Novell, and if not then be brave, drop Solaris and really partner with Linux and M$.
I can't see Sun selling the x86 business as Chris said, especially not to IBM who seem intent on selling their own xSeries to Lenovo. Besides, it's the Sun product line that would seem to actually make a profit whilst incurring the least risk/investment as the majority of the technology is not developed by Sun.
MySQL? Swallow the pride and sell it now, whilst it still has some goodwill value. Maybe Oracle would take it just to stop M$ buying up a competitor, or maybe SAP would like their own database. Either way, I can't see much of a buyer queue forming in a recession.
And a merger? Chris, with whom? HP or IBM are simply unlikely as neither really has the need, and if one proposed it the other would probably start legal/monoploy proceedings just to prolong the agony. Fujitsu seem off the buyers list - apart from the fact they're busy divorcing Siemens, there is the question of do they really want to buy anything from Sun. Remember, Solaris, and SPARC are "opensource", so if Fujitsu really wants to continue with either it doesn't have to buy the over-staffed and over-priced divisions from Sun. Fujitsu already have their own bigger x86 bizz and work with M$ and the Linux boyos, so they don't need anything there either, and a declining Sun customer base is probably not a big enough draw. Hence I am not surprised to see Chris leave Fujitsu - once seen as Sun's White Knight in waiting - right out of the picture.
And the EMC idea is so left field it should have come with a disclaimer. I cannot think of one reason why EMC would want any product from Sun's portfolio other than some of the patents, and with Sun opensourcing ZFS there is no need to buy Sun even if they thought ZFS was worth it. Would EMC pay good money now for the Sun tape bizz when they can simply wait and pick it up as a bargain later?
In summary, Sun needs to concentrate on the x86 business and make it such a core product range that skilled x86 people will want to work there just like UNIX and RISC people used to. But that means jettisoning the majority of the old SPARC/Solaris, MySQL and most of the storage business, otherwise it will simply drag the x86 part down with it. If Sun can survive as a "me-too" x86 business and follow just behind the bleeding edge of the market, then maybe it can survive long enough to look forward to a future where it can again move into being a tech innovator.
Wierd stuff...unhelpful or not?
As a Sun Employee and someone who has had a long term affiliation with the company - I have to say I am not sure that essays such as this actually do the company any good.
I am not saying we (Sun) are in the best position, nor am I saying that all the best decisions have been taken in the last few years - but then again what company can say they have!
I think its the final bit - the summing up within the essay I like the least - seems to suggest that we (Sun) cant continue unless we get broken up....which I think is wrong....and essays such as this on such a respected website just cannot do the company any good! The question must be asked that if the author is a shareholder - is he trying to buy more shares at a cheaper price or make money on his current holding?
Ground level employees and mid to upper level management know exactly whats wrong and where the tinkering (as any company does) has gone wrong.
We are in a different world now - the majority of the major businesses around the world that we deal with are in declining sectors - as all IT companies and IT service companies are seeing.
Yes - we need some more money - and soon - but in this climate show me a company that doesn't - anyone with money in the bank better be getting more to ride out this storm (as a point of note - we do have lots in the bank and the steps we are currently taking aim to keep us that way!).
Sun can still stand up on its own 2 feet - Solaris - is one of the best implementations of Unix I have seen - lets make no quarms about it - why have tools that can potentially harm your system installed by default - why would a system running a webserver need a whole set of compilers available to it? It doesnt - that sort of craziness just creates security holes - so we dont install by default - if you are that lazy to install the dev pack each time - create a jumpstart image or a flar or a vmware image or whatever your flavour of quick install is!)
Our hardware of recent is 2nd to none - and hell - we are even windoze certified!
Have you ever tried using a SunRay? I can go anywhere in the SunWorld (even @home on a broadband connx) and get my session from work using my ID javacard - and though performance for multimedia presentations is limited - if I want to compile something - read email - surf the intranet - work on documents do just about anything I would do at my desk - I can - and have it instantly available to me no matter where in the company I travel to - the performance and licensing constraints are far far more favorable than a citrix dumb terminal or any other thin client implementation I have ever seen!
To sum up - yes we are getting a little sticky here - but less so than most of the banking system and definitely less than MFI or Woolworths! Sun will be around for sometime to come yet - as long as people do not try and talk us down and create bad feelings about the company to send jitters to the markets - this is precisely what happened to the banks in the last few months - why do this to IMHO one of the best OEMs around - and thats not just because I work for them and hold shares but because I truely believe that the company tries to do "the right thing" and not just to "make money".
questions and answers
Question: "How does having a cash pile equal to market cap make the company worthless."
Answer: Yes as an "ongoing company" shut the doors, fire everyone you cash out with more money in the bank. Obviously the shareholders whould rather sell parts of the business keep a subset and have their share value increase. Has Sun ever thought about giving a dividend? They wasted $3B in stock buybacks only to see the stock drop 85%. The only thing the stock buybacks did was rewards the shareholders that sold out. IBM and HP give billions in dividends to their loyal shareholders.
Question: "You only have to look at the share price of IBM, HP, MSFT and others to see this and Sun is no different. Sun unfortunately was still in the process of trying to recover its stock price when this downturn hit and with all the short selling of stock that happens these days (should be illegal really) it really give a false impression of how a company is performing."
Answer: click on the following
That's "jake". The other Jake is a different person. Ta.
"You're so dead wrong. The hardware is good, that's true, but it's in software that Sun is not only leading the way but in fact blasting away all opposition."
Look, I've been using Un*x systems for thirty years, give or take. I have Solaris 10 systems installed in Fortune 150 companies. I have BSD systems installed in Fortune 150 companies. I have Linux systems installed in Fortune 150 companies. A tweaked Solaris 10 is a bloated piece of shit, compared to stock BSD on the same gear. A typical Linux install is somewhere in between. Tweaked BSD and Linux systems are (typically) about equal, and both run rings around Solaris.
"Without them, we'll be back to the mindless mediocre feature tweaking of every other OS producer out there."
In other words, you believe marketing hype and/or aren't really a sysadmin. Anyone who thinks a swiss army knife OS installation is better than a carefully targeted system hasn't been in the business long enough.
Trying to be all things to all people is a mugs game. See Microsoft.
Over 1000 words in your latest essay; really, I don't think anyone is listening anymore :)
Twatt aside, while I generally enjoy the El Reg Sun articles written by Timothy (and previously Ashlee), I have to say this one was pretty lame...covering ground a lot of other tech journos went over, ohh, three or so weeks ago? "Oh noes, Sun stock is tanking, let's play a guessing game as to who is going to acquire...!!"; slow news day p'raps?
For the benefit of your readers, some rather more rational POV is provided here:
Which as far as I am concerned is much closer to the reality.
Some refreshing perspectives on Jonathan's performance:
Schwarz is, what, 42? The very strong vibe I get is that retirement is not on the cards! WIth the latest Wikimedia win (along with FOX and FAA), we are seeing the beginnings of his strategy coming to fruition.
What do their customers want?
I've been buying and recommending sun kit for two decades but Sun simply refuses to sell us what we want to buy. I watch the required patches to Solaris 10 and I don't see a production product yet. I have solaris 9 machines with years of up time that have all the needed patches. While dtrace, zones and zfs seem nice, we don't need them but we need stability and we just don't see that when we have to reboot for patches every few weeks. Our preferred platform is currently the netra 210 because they will run solaris 9 and they fit in our old racks but its last order date is approaching. Once we can no longer buy them, we will be switching to bsd on some other platform.
Same old, same old
The problem Sun has today is the same one they've had all along. They're ignoring the bottom-end, entry-level, mindshare-gaining sector of the Unix/Linux customer base. Sun is addicted to the high-end margins of proprietary hardware and all of their problems come from that. It's really that simple.
Compare and contrast. Sun developed a great product with Sun Net Manager just a little after HP's Openview. When Sun realized SNM wasn't selling as many boxes they cut its budget. HP, OTOH, was into OpenView for the long haul, and not to sell boxes, so continued developing it. Today OV is entrenched whereas SNM is forgotten. It was the same story with Mailtool except that product was far superior and far ahead of Exchange. Ditto PC/NFS, OpenLook (even though it was ATT who dropped that ball Sun could have run with it), SunOS4 (the BSD code base that Sun spent so much energy switching from, to get to SVR4)... Apple leveraged the same BSD to its current success and Sun's fortunes have fallen in pace with SVR4. The only reason Sun is still around is Java, which only ran well on Sparc for a few years, and has had its development budget cut for the last few years too. It's an old story, repeated often by companies from 3com and Novell to Yahoo.
It is really too bad the hardware-centric directors and bean-counters have such a strangle-hold on the company, and none of the vision that propelled Sun in its early years (when mailtool, calendar manager, SunOS, sunview, ... were years, many years, ahead of the competition).
So McNealy, Schwartz, where's the vision?
> I am not sure that essays such as this actually do the company any good.
What gave you the impression that the authors intention was to do any good? Rags like the register thrive on scandal and big news bad stories - A Sun breakup would be great for the register in the same way that a royal divorce or death is great for a british red-top newspaper.
> on such a respected website
You are joking right?
> as long as people do not try and talk us down and create bad feelings about the company
Hmmm - you mean like everyone did to HP a few years ago - didn't see too many Sun employees holding back then (certainly not those at the top anyway). Maybe you personally didn't join in, but to expect the media and your competition not to try to take advantage of the situation is naive in the extreme.
Sun technology I will be the first to admit is generally great stuff... Sun the company though? Not sure the same can be said.
Market capitalisation is the total number of share times the share price - effectvily what the stock market thinks the company is worth.
Suns "cash or cash eqivalents" effectively money is hte bank is currently almost exactly the same as the market capitalisation.
Think about it the stock market which to a large extent conists of people who value companies for a living has collectively decided that the only valuable thing about Sun is its bank balance , it values the product line, research, customer base at zero.
They may of course be wrong, but Sun is slow to produce competetive chips, and its management have bet the farm on being a software company. Except they have no history of succesfully SELLLING software although they are very very good at giving it away.
My moneys on either Fujistsu which really wants to be a credible hardware company, or, IBM which is making a pile of cash out from Java/J2EE and may want to take control of the technoligy which is a core part of this cash cow.
Regarding these comments
"ZFS, DTrace and Zones/Containers in particular represent an order of magnitude over what went before, and over what any rival vendor is developing."
"AIX is rock solid and very manageable but lacking any real innovation"
Have you actually used AIX in the last 10 years ?
Its a little more complex than people seem think
Sun have two main problems, as they are an engineering company they tend to get driven by technical innovation and as a consequence, products don't ever seem to get that final 'polish' that their competitors have on their products or they drop them become they come to fruition (the Solaris patch tools, Sun Management Center, PC NetLink are good examples).
Secondly, some of Sun's products are so innovative that customers don't 'get' them. For example, the T1 and T2 servers make fantastic web and DNS servers (as they are low power, have integrated crypto, come with LiveUpgrade, etc) but as customers are so obsessed with clock speed, they miss this fact. Considering that a single quad core Xeon based machine can run enough apache instances to flood most commercial hosting lines the T2 based machines make huge economic sense in terms of power and operational savings alone but its a fact that is missed by most sysadmins and IT managers; something that Sun can't really do much about.
OEM staus of HP and HDS
correction : HP doesn't OEM from HDS but direct from Hitachi itself just like HDS does, which means HP buying the Sun storage business might not benefit HP at all - it'd be an OEM buying from an OEM an OEM'd system :-)
"....http://blogs.computerworld.com/sun_revenue_loss_q308...." Nice puff piece, maybe Arun Taneja got asked to turn a favour for his old buddies at Sun. I can't see Jeff Byrne, who has much more experience than Jeff Boles in how the large vendors work, writing such twaddle as "....what makes me think Sun is well positioned - I am hard pressed to pick out visible technologies that are not well aligned..." - if Sun was so well-aligned its sales wouldn't be in such a nose-dive. And pretending it's just for this economically slow period is hilariously blinkered or just downright dishonest - how many years is it since Sun had four profitable quarters in a row? Compare to HP or IBM, which can both answer quickly enough. And maybe Mr Boles would like to compare notes with Gartner or IDC, both recognised names in the market, who both disagree with Mr Bole's point of view. I don't even have to guess which one customers would value.
"....http://www.redmonk.com/jgovernor/2008/11/26/what-should-sun-do/...." This is a great article for Sunshiners, the rest of us simply don't believe lines like "...The retooling of a tired mess into a clean and solid portfolio is complete..." - complete!?!?!? Hold on a sec, weren't you Sunshiners saying it would be complete when you got Niagara? Oh, but then you had to buy in a low-end SPARC64 design from Fujitsu because Niagara didn't fill the hole at the low-end, and you're losing market share to Lintel/Wintel, low-end Power and low-end Itanium. Hold on a sec - even with the M3000 you're still losing at the same rate, and now just have a mess of non-compatible Solaris versions and overlapping chip designs! And weren't you telling everyone to skip the Mx000 series and go to Rock? Oops, no Rock! Suddenly the Mx000 series is just fine. Laughable! And don't get me started on the Sun blades, which are the most comic offering in the blades sector. As above - if Sun is so "complete", why are you redlining, and why have you been redlining for so long? All in all, the blog is a nice quote from an unknown, hope the memory keeps the Sunshiners feeling loved as they wander down to the dole queues.
"....http://www.cuddletech.com/blog/pivot/entry.php?id=988#comm...." Hilarious! The best post you can find is a piece admitting all the problems Sun have? "....In this tough economic climate I don't think anyone has the time or money to take on the problems of Sun...." Says it all - if Sun was such a great buy, it would be bought, if just for the cash in the bank. The fact is NO-ONE WANTS TO BUY SUN because it would be like buying a diseased sheep - you can't sell it on, the meat is unfit, you don't want it to mix with your herd in-case it infects them and reduces their value, and even if you skin it you may not be able to sell the fleece. Ben Rockwood may think Schwartz is doing a good job but note how he deflects blame to the board that appointed him - this is just silly, as that means the board makes bad decisions which somehow alleviate Schwartz of any blame for his continued operational misfires, yet they should take the longterm blame? I'm sure that Ben doesn't apply the reverse and congratulates the board when Schwartz does anything right. Mind you, Ben can't get stuck with that dilemma too often!
Oh, and I see from Ben's blog you have taken to posting under my name further afield too. Amusing to see that you find a little humourous needling such a threat you have to run all over the web posting rubbish under my nom de plume. I think a better use of your time might be learning Linux and applying for a job at IBM, HP, Novell or Red Hat, they might have openings for a semi-technical marketing droid.
Don't flatter yourself Matt B
I'm sure Matt Bryant is a perfectly common, normal, and boring name...there's bound to be other people out there in the big wide world who might actually have a different POV of Sun than your entirely singular perception, who might even feel perfectly happy posting on blogs...
Gartner results just in.....Sun down double digit again...4Q will be a complete disaster
Expect Sun to be sold in pieces by the end of the year so IBM or HP(SPARC/Slowaris) and EMC (STK) can write it off in 4Q.
Obviously it will not be Fujitsu is also in a freefall and already spun off their chip business.
"Sun came in fourth when ranked by sales, with $1.16bn in boxes sold, a decline of 13.7 per cent from the year-ago quarter. The Fujitsu-Siemens partnership (soon to be just Fujitsu since the Japanese company is buying out its German partner) accounted for $616.7m in sales, down 7.3 per cent and obviously falling faster than the market at large."
If you look at the subtext of the engineering direction for 2009, they're gearing up for a crack at the general (rather than media) mid-sized enterprise market.
They have successfully productized and monetized an open-source Unix and could easily absorb the best of Sun's R&D people. They're already cherry-picking bits of Solaris (Dtrace, ZFS &c). Apple could give Solaris the makeover it deserves.
They could do with a decent x86 server series. But they've never been afraid of using non-Intel processors, either.
What would Apple do with Java or with StorageTek, though?
RE: Joshua Goodall
Apple is an interesting idea, but probably not too realistic. If we are going into a recession, manufacturers of luxury goods will suffer as customers save money by buying budget brands. Apple is very much a luxury consumer brand for the majority of its range, so I expect Jobs and Co to be carefully hoarding cash for the coming lean period, and not investing in a whole herd of white elephants.
"....If you look at the subtext of the engineering direction for 2009, they're gearing up for a crack at the general (rather than media) mid-sized enterprise market...." They don't have the channel presence reqired, unless they try the old Dell sales model and go only online. This has the downside of zero feet-on-the-street, and nobody out there talking customers into trials. If they buy Sun just for the channel then they'll be disappointed. And if they did they'd have to service all those existing Sun SPARC/Solaris customers, a phenomenal drag on any buyer. If Apple were to want anything from the Sun firesale they'll wait for the bargain clearance rather than paying the type of over-inflated price the Sunshiners will want. And what do they need to buy? They already have very skilled x86 engineers quite capable of matching Sun's designs. So maybe a few Sun refugees will end up at Apple, but I can't see Apple making a home for the lot of them.
"....They have successfully productized and monetized an open-source Unix and could easily absorb the best of Sun's R&D people. They're already cherry-picking bits of Solaris (Dtrace, ZFS &c). Apple could give Solaris the makeover it deserves....." True, but in a narrow niche of the desktop, not in the enterprise space. Making and selling desktops is completely different to even Wintel edge servers, let alone real enterprise servers. Currently, as I understand it (feel free to correct me if I'm wrong as I am not a MAC OS guru), MAC OS doesn't scale beyond sixteen cores, and Apple have no experience in making multi-socket servers with mroe than four CPU sockets. It's a bit of a leap to think they would suddenly push out a new multicore OS and server design in short notice without some info leaking or being leaked to the press. And they wouldn't want any of the current Sun designs as the real ones are Fujitsu's, not Sun's.
"....They could do with a decent x86 server series. But they've never been afraid of using non-Intel processors, either....." But they already have a skilled engineering team that have Intel experience, so why pick up Sun's, especially when it comes with all the baggage of the SPARC and Solaris business.
Since the FSC divorce broke, Sunshiners have been desperately looking for a new white knight to replace Fujitsu, and several have minetioned Apple because they see synergies in their "non-conformal" designs and because they faltter themsleves that Sun and Apple are "innovators". Unfortunately, it's a bit like expecting Ferrari to buy a company that makes three-wheeled delivery vans - yes, they both are different to the rest of car/van makers, but that doesn't make them in any way alike enough for it to make sense for Ferrari to buy them.
Just one piece
The question “who needs Sun?” is not the new one. Before do an answer take a look on how Sun's body could be served?
4. Fresh cuts in STK!
5. and Mysql.
Did I miss anything?
Let's reorder more carefully:
1.Solaris and Sparcs are symbiotic bodies, if you divide it – they will die both, so it is really a customers base
4.Mysql - something smells phishy with $1B price tag
Lets list the bidders
1.IBM, HP will love to replace everywhere any remains of item #1, so assume, no bidders at all.
2.Java – is a community and it is too hard to buy a community. Let's assume IBM just will hire James Gosling.
3.STK - EMC or HP
In the result, the only one piece of Sun could find any bidders – STK. At least, it has some value and could be detached from Sun relatively easy.
The rest of Sun will stay as it is and will continue to Rock the Sparc roadmaps and release Mysql and Java updates.
This seems like an odd article...
Considering SUN is an Open Source software and hardware company, there is really not much to buy except engineers... it does not make much sense to me that anyone would buy them since an open-source portfolio is a "poison pill" - if you want it for producing a product, you participate in the product with a head-count, and you can use the technology.
The product road maps have been very good for the past couple of years, including the T1, T2, T2+, VI, VII processors. There is a lot of talk about people being disappointed in the release of another processor, but plenty of high performing SPARC processors seem to be making through engineering to meet business requirements. Considering SPARC has always been an open platform with multiple chip vendors, I don't really understand why people are so caught up over a particular processor when there have always been plenty of others to take up the slack from various vendors.
After working on Solaris 8 & 9 for years and using Solaris 10 - I am uncertain why anyone would really want to go back to any pre-Solaris 10 operating system. People can even run Solaris 8 or 9 in a container under Solaris 10 and migrate off an old system entirely using BrandZ and get the benefits of wire-speed crypto engines, extremely high-throughput 10 gigabit ethernet, multiple virtual servers, and huge error correcting ZFS volumes to keep data on (vs the smaller logical units supplied by other operating systems.) For consolidating older platforms where increase in thread speed is not important, CoolThreads makes the mark.
Considering SUN published the highest throughput single, double, and quad socket platforms on the market for a good portion of 2008 and the most advanced operating system features are in Solaris OS - the only questions are market penetration, cost, and when the next speed-bump will take place.
Will the market support the costs to get the best in class or is the market willing to wait 6 months for the next catch-up? (Intel came out with a very interesting i7 series, for example, albeit bottlenecked on an internal bus throughput for some applications, but then again, some people will rightly be concerned about an odd application here or there which experience the single thread bottleneck under CoolThreads.)
Since SUN now sells proprietary Intel & AMD with volume pricing advantage and SPARC - Sun appears to have mitigated the leap-frog problem, releasing both Open Source and the fastest processors on the market for the past several years running while selling the proprietary Intel/AMD to make up the difference during those periods where the market place catches up with the option of superior scalability in SMP platforms in OpenSPARC to hold the high-end.
The real issues don't seem to be addressed in this article for where SUN is being negatively impacted... this centers around proprietary software licensing per core, negatively impacting the software costs in an Open SPARC and Open Solaris commercial implementation.
The move by SUN to forge ahead in Open Software (i.e. Postgres, MySQL, JAVA, etc.) to mitigate the proprietary software vendor negative licensing impact on SUN has been a strong counter, but until vendors support MySQL along side Oracle on commercial software - SUN will continue to lose commercial deals when systems are priced if the systems are priced do not include some of SUN's unique hardware capabilities.
no more bright ideas
If ponyboy sells of the hardware divisions of Sun, then Sun is dead. The time is now ripe to make money in opensource software and operating systems (yes you can sell OSS), people are dissatisfied with MS and Vista and their endless frameworks, Apple has never been a friend to the unwashed masses, but Sun is doing nothing. Now is the time Sun needs leadership, and ponyboy simple opensources everything so that 'all boats will rise' (his words). In the past Sun has had enough bright people working for them so that they always came up with ways to make money, but back then they had bright people in charge. But now they are floating aimlessly in the middle of the OSS sea, and their boat is not rising. Sun will sink, they will get bought out cheap, and that will be the end of Sun.
ROCK is coming...
"Rock is hot on the heels of what we just announced and will bring to market some new techniques to aid in parallelization, aid in the effort of programmers to write parallel programs with transactional memory built into the hardware," Hetherington says.
"We've got systems that will be getting out to beta sites in the very near future. It won't be long before John Fowler is back on stage to do another announcement."
Keep it up Sun!! Opensolaris 2008.11 today, Java FX on the 4th, Solaris 10 u6 in the bag, an afforable (hopefully) T3120...and Rock to kick off the New Year...
ROCK is coming...
at a time when the other UNIX vendors will be shipping Tukwila and POWER7. But it is hard to know. I mean the messages given from SUN was that it would arrive in late 2006, early 2007, then it became 2008, then mid 2008, then it was 2009 and now it's late 2009
I wonder how a chip originally scheduled for 2007 will fare against the competition. I mean there is big difference between competing against POWER5+ and last iteration of POWER6 or first generation of POWER7.
ROCK is coming...
My guess, why SUN has a delays with ROCK is a technology. Sun wants a high clock and relatively a low power on coming Rock, but TI stuck on 65nm and that is why Sun moved to TSMC to catch up 45nm folks (IBM, Intel and AMD). Currently, TSMC is hoping to unveil their 32nm process in Q4 2009, that is why Rock is directed to late 2009.
From architectural point of view, Rock will be very unique and it will have some advantages versus processors from Intel or IBM, no issues here, but ... the time gap in the delivery to the market of Rock will play the vital role for the whole Sun's CMT processor initiative.
Rock is coming...?
Reminds me of the old Spitting Image skit of David Steel and David Owen, where they are wating for the Liberal Wave to Rise and sweep them into power, and David Steel's puppet keeps ecstaticly wailing; "It's coming, David, the wave is coming!"
Of course, the wave never did come.
Just for the Sunshiners with very short and selective memory, UltraSPARC V was canned after it had taped out, after "beta" units had gone to seed sites, and after Sun had assured us all it was "revolutionary" and would be better than all the competing chips, etc, etc.
/Laugh, point, yawn, laugh some more.
Hello....The Rock team has been canned
done...gone....history....Rock will never be announced let alone be available in any Supernova servers.
The transactional memory did not work.
The instruction cache and data caches where shared by different micro-cores
Oracle is too much of the server business to introduce a chip that requires 12 licenses * $47K
Server revenue in 3Q was down 17%....6,000 job cuts cannot avoid an unproven chip
Niagara III will survive thought...
Oracle on a 12 core chip...
An Anonymous Coward suggests...
"Oracle is too much of the server business to introduce a chip that requires 12 licenses * $47K"
That is why SUN has Containers & Zones with the ability to cap resource.
Run Oracle on a couple of cores, reduce the license costs until more capacity is needed, buy more licenses as required, and run a command to increase the number of cores used by the Zone.
Re: Rock is coming...?
You still waiting on Whitefield? Here's a hint... It was canceled.
How 'bout the fact that Tukwila was promised to be ready in 2007? Hmm, how's your memory doing there little Matty. Still waiting for Tukwila? You may have been drinking too much of the HP and Intel Coolaid. Support for Itanium is dropping faster than little Matty can write a 1000 word essay on why Sun sucks.
Montvale was a year late as well, by the way, with several "important" features dropped and initial speed lower than originally promised (about 20% less).
Come now little Matty. Even you have a better memory than that. Anyone espousing the perfection of HP and Intel cannot throw stones about canceling projects or being late.
No need to point and laugh at little Matty, though... I have sympathy for the memory impaired. Besides, I feel bad for Little Matty as he will be spending a lot of time pretty soon porting everything over to Xeon when Itanium is canceled all together. Or will Intel play the name change game again and rename Xeon to Itanium? Those crafty marketing geniuses...
IT? because I doubt more and more every day that little Matty even works in the IT field... and I don't count HP sales as working in the IT field...
"Oracle on a 12 core chip... "
Oracle on a 16 micro-core chip requires 12 Oracle licenses.
That is $688K in the first year with maintenance.
Multiply that by the 8 chips in the largest Rock system is $5.5M.
Oracle does not recognize LDOMS for sub capacity pricing so Sun customers are forced to try to use zones but $344K for half of a chips performance is still insane.
All this is mute anyways since Sun has already canceled Rock.
Oracle on a 12 core chip
I think what the anonymous poster meant was that if Rock is 30 times faster than the 1.2GHz UltraSPARC III. Then it'll still only be 2.5 times faster per core. Now that isn't very impressive thinking that for example the 4.7 POWER6 launched in August 2007 is almost 6 times faster than the UltraSPARC III, on specINT 2006. And I would imagine it's even worse on DB OLTP loads.
So when ROCK arrives if it's in late 2009, then it will be facing for example a power7 almost right from the start. And well I guess that POWER7 will be faster than POWER6 on a per core basis, so you will STILL have to pay 3-4 times the license cost to for example Oracle for the same performance, as you will on SUN's fastest competitors.
Which for a workload that corresponds to one ROCK processor would be a difference in 235KUSD-282KUSD, 8 versus 2/3 licenses. And on 4 sockets this would exceed 1MUSD.
Matt Bryant is coming...?
Reminds me of the old Spitting Image skit of David Steel and David Owen, where they are wating for the Liberal Wave to Rise and sweep them into power, and David Steel's puppet keeps ecstaticly wailing; "It's coming, David, the wave is coming!"
Of course, the wave never did come.
Just for the Itty Bitty Machiners with very short and selective memory, T1, T2, T2+ were released ahead of time and performance on-par or better than promised. Sun had assured us all it was "revolutionary" and would be better than all the competing chips, etc, etc.
For the better part of a couple of years, application performance was best in class in 1, 2, and 4 socket machines.
/Laugh, point, yawn, laugh some more.
To Anon Corward - "Oracle on a 12 core chip... "
An Anonymous Coward says, "Oracle does not recognize LDOMS for sub capacity pricing so Sun customers are forced to try to use zones"
"Hard partitioning physically segments a server, by taking a single large server and separating it into distinct smaller systems. Each separated system acts as a physically independent, self-contained server, typically with its own CPUs, operating system, separate boot area, memory, input/output subsystem and network resources. "
"Examples of such partitioning type include:... This is not a comprehensive list of all the different types of technologies or resource allocation devices/programs that would fall into the category of Hard partitioning."
If capped zones are acceptable, LDOM is obviously acceptable, since CPU capacity is added a thread of a core at a time, has it's own OS, separate boot area, memory, separate I/O subsystem, and network resources.
All the requirements are clearly met, by the letter of the Oracle licensing law!
- Crawling from the Wreckage Want a more fuel efficient car? Then redesign it – here's how
- Review Xperia Z3: Crikey, Sony – ANOTHER flagship phondleslab?
- Human spaceships dodge ALIEN BODY skimming Mars
- Downrange Are you a gun owner? Let us in OR ELSE, say Blighty's top cops
- Ex-US Navy fighter pilot MIT prof: Drones beat humans - I should know