Here, with the stock market melting, we find Sun Microsystems in most uncomfortable territory. It's got a stock market value of $7.7bn, which means that the one-time lord of the servers is a mid-cap company. Our friends on Wall Street warn that Sun needs to maintain at least a $10bn market capitalization to stay in the large cap …
Where's the Sun girl?
Sun has all the bits to really shine in the tiny computer market, except that end user gizmo.
Combine mini semi-thin desktops that feed from big Sun iron along with wireless armtops and tablets and it could leverage its software investments in OS, office and Java and deliver very sharp price points along with solid engineering.
All we really need is that pic of the Sun Girl getting an all-over tan while working on her spreadsheets on the beach.
Personally I've always found Fujitsu SPARC gear to be better designed than Sun's gear. Maybe Fujitsu should just accquire Sun and turn them into a software-only division. Then just slap a "The Sub company" sticker un the Fujitsu logo on the hardware. In short:
1) Fujitsu buys Sun
July 10th is the day when company is going to issue pink slips.
Sun never figured out the SME
If you're a government, or a government contractor, and money isn't an object, then of course you buy Sun's. For Sun, its good work if you can get it.
If you work for a smaller organization that has a real budget and actually looks at lower cost alternatives, there's absolutely no way you would buy Sun. They make Apple look like a value retailer. They charge thousands of dollars for add-ons like hard disks that you can buy off the shelf for hundreds of dollars.
It amazes me that they're in business at all. I understand that the really big guys like them, but even when our medium sized org looks for high-end workstations and servers, Sun is ridiculously overpriced. I've been tempted by Sun, but never able to justify paying for anything they sell.
So yes, they innovate, and they have software that is pervasive, and... they just can't figure out how to sell anything to the rest of us.
Fujitsu's shareholders would do well to remember Unisys. If they must spend their money then the IP would be worth something but not at the cost of acquiring and then 'letting go' 30k employees.
They did themselves out of a market
The damn things don't break. About half the small to medium companies I have had dealings with (in France and Ireland) have a sun server buried away somewhere, usually bought as an application server sometime in the late 90's and still dogging away as a file or database server. The last one I saw was 11 years old and, in a fit of madness, had been beeping to its self for months. A dusting off of the terminal showed log files overflowing with temperature warnings, and a quick once over turned up a seized power supply fan. I would bet my socks the damn thing will still be slogging away in another 10 years. Too good for their own damn good, it would be a real shame if sun where bought out but it does make a lot of sense and would be infinitely better than loosing them for ever.
'twas a few years ago that my 2 CPU 400MHz Xeon box outperformed a new dual CPU Sun box by a factor of 3 running SETI@Home.
I know that's probably not a fair test, and if only M$ had released SQL Server on Solaris it might have been a much different story, but then I have heard it called Slowlaris.
But at the time it led the senior IT manager to exclaim "I'm not going to buy any more of that expensive purple shit!"
Aah !! Sun !!
I've always had a soft spot for that company. At a time when the others, now extinct, like NCR, Data General, etc. were trying to bully their way into any business going, the Sun people will calmly explain the techie aspects of their kit !!
I have always had utmost respect for their kit and their software. However, it's a shame that their marketing people got their pricing so wrong !! Some more realistic pricing will help their cause no end !! Having said that, their support engineers and people are wonderful most of the time !!
Outlasting the four horsemen
I think Stan's hit the nail on the head, Sun kit is just well 'ard, I still have an old SS20 pizza box with add-in Ross CPUs and a casing that's taken more beatings than Templeton Peck, and the damn thing still serves up shell and DNS over 10 years later.
Don't let the Sun go down...
These are one of the few guys who are actually still investing in the Unix and non-x86 areas. I don't want them to die, and their hardware is definitely the best in the mid-range/high-end area, second only to mainframes. I'd hate to see them go down! And as Ishkandar just mentioned, they've always been good on the techie aspects.
I've worked on their E25k boxen, and those beasts won't go down unless somebody does something really bad on the hardware. Uptime/availability on those mammoths are something any Wintel server would only dream about.
Another example of El-REG stupidity
Just replaced a metric buttload of desktops. I had a choice between Sun Ultra-24's Quad-Core with 8gb memory DL-DVDRW and 250gb disk at $1000 or a Dell Core2Duo with 4gb memory, DL-DVDRW and 160gb at $1200
Obviously I went with the Dell because Sun's cost too much and besides Sun's are just so unreliable and under performing I'd have to replace in another 10 years unlike Dell/HP/IBM which will fail in a year or two
Maybe the arcticle and following comments explains it all
As an Ex-Sun employee (still have my Sun shares) now working for an end user this article and the comments following it says it all:
- Sun is not seen as a company that can deliver product which suits would buy - the techies love it, it makes brilliant kit but somehow the buyers are not getting the message.
e.g. Intel changes its products all the time, all x86 servers is already superceded by the time it is delivered but the lemmings are still following its marketing message ...
Hope the Sun marketing droids are the first to get shot - firing the sales guys when the marketing message fails is just stupid. Sun has no clue what the market wants - the days of technical savvy people buying your product is gone - sell cheap badly designed kit - that is what the end users want - well engineered masterpieces is so yesterday. And the blisters on the fingers of the analysts who talked it up before the bubble is still smarting.
PS: Beelzeebub - your comment with regards to a suit making a buying decision based on an opinion by a wanna be techie (that is you) is so true.
But it's the Age of the Freetard!
Which means My LIttle Pony now has to give the company away for nothing, right?
OK - so slash the R&D, give away Java and Solaris, make loads of mediocre midrange boxes... is there anything he hasn't done yet?
And you'll read it on his blog first.
I see your Schwartz is bigger than mine
Meh, I'm a recent Sun convert and I got a good feeling about the future. It was actually through Apple (ZFS + OS X) I got to know anything about who they are and what they did, prior to that I didn't really have a clue. After years of tearing my hair out dealing with Apple and Wintel+MS kit (well, not so much the Apple kit, but certainly the latter!!), how refreshing it's been to dabble in the products and culture of a completely different company.
Thanks to Sun's openness, as a UNIX newbie wanting to leverage myself out of my desktop support role, I've been able to get my hands on some amazing products and technologies which either have no peer, or, would be impractical cost-wise to try any other way. Can't think of any other company where this would have been possible. Just about to purchase my first entry level server to run my projects using Solaris 10 + ZFS, and Sun Ray Server Software - and it'll be wearing the Sun badge on the front, for sure.
In spite of all the bitching about Jonathan's management and his blog etc I think image-wise he's valuable to Sun. He's a generation removed from the Jobs, Ballmers etc, certainly more approachable and charismatic, with a real sense of humour to boot. Jobs is an autocrat, Ballmer certifiably insane...and other big tech CEOs to end users such as myself are pretty much invisible. I think having someone as comparatively young and dynamic at the helm gives the outfit a positive vibe. I really feel the company is going exciting places, and genuinely doing exciting stuff. I don't get that vibe from HP, Dell, MS etc. At all.
Certainly when looking at the Sun Ray stuff for example (marvellous!) it's surprising just how many of my colleagues in senior IT roles are blissfully unaware of its existence or the company that makes it.
Is Sun burning out?
Sun needs to pull its head out and listen to what its older customers need. I want servers that fit in my racks (which were bought to host Netra X1) and I will not run Solaris 10 as its too broken and I can't audit its security on a lean system (think 4 packages on the system in total) like we can with Solaris 9. Who had the bright idea to put startup/shutdown data that has to be audited in a proprietary binary format? Its supposed to be Unix and with modern hardware there is no time savings with binary files even if they are parsed a billion times. I buy Sun for all Internet facing servers since their nice little hardware stack provides an extra layer of security against buffer overflow problems. We run quite a bit of Sun hardware ranging from antiques like Sparcstation 20 and SS1000 to the newer telco grade Netra 210 which seem to be built like a tank.
One can't blame Schwartz, really. I have to agree with Ishkandar and that one AC on the marketing. BMW and Porsche, they have marketing. And people are proud to own one of those cars. Even Apple are able to sell overpriced stuff to hordes of people. Only Sun cannot sell performing stuff at relatively reasonable prices to people.
Plus, they have been and are still screwing up on the FOSS-front. Yes, Simon Phipps, we've been through this up and down. And still you don't get it. FOSS is community. Not community controlled by the whims of Sun.
As long as the chain of command at Sun decides is a 'community' edition is made public, and when, it will not work. (For those not in the know, SXCE is the 'community' edition; but not all are made available to the community; e.g. 88, 92)
It would be a real pity. The H/W is hot stuff. Those things get punished during testing, I'm not surprised they last so long. I might buy a Sun desktop next, and OpenSolaris looks interesting. Solaris has some clever stuff under the hood.
They are clever people, though I had qualms about McNealy's legal wranglings with MS, that hurt Java more than it hurt MS.
The worst thing about Sun dying would be that they're nice. They bent over backwards to keep employees. They give the prongs to Wall Street. How many companies do that?
It would prove Trump, Ballmer's, Gates mantra that only bastards succeed by bullying their way around, regardless of employees and, you know, R&D and the desire to make better products than the competition, rather than throwing their weight around. Sun are like the reverse of most companies - good products, bad sales and PR. The others are shovelling shit that's dressed up as The Next Big Thing, while Sun have got the crown jewels. It's just that nobody's heard of them.
Please stay alive Sun, my faith in humanity is at stake!
Hey, I make the offers to our clients... and while they like Sun stuff... they don't like their prices.
Yes, they may last a decade... but what is needed is faster and cheaper stuff... even if it breaks down in 4 years or so.. that is the reason we have high availability... and no, HP stuff doesn't break in 4 years as a yugo....
A company that's had it's ups and downs over a long period. Pity this down has arrived during the current economic situation - but, let's be honest, has any established player's market value held steady, let alone improved greatly, this year?
They may not be bleeding edge, but their products seem both stable and long term reliable. Personally, I can see them pulling through, so long as a predator doesn't pick them off and wreck them.
Be a shame it that happened. Sun have been pretty much a bench mark. Kind of like the old days when 'no one got fired for buying IBM'.
The old, old story of commoditisation
This is the old, old story of the commoditisation of a product. It's been killing off the specialist computer companies one by one, and only a very few niche players and those who can adapt to the pile-them-high, sell them cheap game survive in the hardware and volume software business. For those that do the numbers, and compare the price/performance of a commodity x86 architecture server with a SPARC one then it tells a story. The Niagara based servers can compete in some part of the market space, but the slow single thread speed is a major handicap on many sorts of applications. It's also clear that Intel aren't exactly staying still either. You have to go a fair way up the SUN product line before you can't find an x86 server which will compete on performance by which time the price/performance gap is huge.
At the very top end, there is some room for really big iron stuff for applications that don't scale over commodity servers. However, there's not that much room as the cheap iron gets faster and faster. Also SUN are not currently at the top of that particular tree either.
So the writing is on the wall, especially for a company that hasn't exactly got a great presence in systems integration and professional services. If they can join with Fujitsu and compete on more level terms with IBM and HP then maybe that's there best hope. It's a bit sad in its way, as it's the last of a breed of companies with a bit of individuality. In this business, you either adapt and change into something completely different (as HP have) or you die or get absorbed which is the fate of the vast majority of hi-tech companies in the IT sector. I suspect the half-life of such companies is way less than a decade.
One day somebody will make a comparison of SUN with Apple - of course they are in very different market segments, but they both share an almost evangelical following of almost religous fervour.
Re: suggestions 4 a turn-around
Not sure if linking to my blog will get through the editors, but i will specifically link to the post I made ab/ the article from Ashlee:
these are tough times, and perception means a lot, but there are ways to get going, without selling to Fujitsu (or Oracle?)...
all the best to those that can see the role that Sun plays...
Not just the hardware
I have binaries compiled in 1993 (Solaris 2.2-ish) and they still run flawlessly on Nevada ("Solaris 11") on current processors. Let's see Windows or Linux give you that sort of future-proofing...
What really matters is......
This is the best example of what Sun is about, the server below needs to be decommisioned, it's unused and insecure to say the least but we're leaving it up as long as possible to see how far it can make it!
Take it as you like in terms of trusting me pasting in text, I've not edited the uptime figures but have removed the fqdn details for obvious reasons.
# rup guardian.#######.##.##
guardian.###### up 2316 days, 10 mins, load average: 1.00, 1.00, 1.01
That won't mean a great deal to the average purchasing guy looking at the price sheets, vendor claims and pretty pictures but probably does to the rest of you.
(and yes, its a ss20 & a credit to our power feeds)
Ain't no sunshine when you're gone
What really lays at the base of Sun's problems is it's naievity, it's naievity and the small size of the company (compared to IBM,HP and Fujitsu).
It's very unfortunate because it's that same naievity and small size that makes Sun so very valuable to the IT sector and both will be lost once the company fuses into a bigger one. I feel ill too, at the very thought.
You can do half a billion worth of R&D and create great products only to find
that 10 million of PR just isn't enough to advertise it, especially not when competing companies love to spend 20 million of FUD stomping your product in the ground.
Sun isn't expensive, but it is perceived that way. Sun isn't 'proprietary', but it is perceived that way.
I were a developer once, Assembly and C mainly and was the first Windoze programmer in a very well known org.
I am now an architect, meh, you don't get any thanks nowadays!
The more you know, the less you know so it seems.
Sun is suitable for all requirements where it is the best fit. But then, other boxen fit most requirements (Itanic not included).
is actually one of its biggest problems.
The number of times I turn up at a customer site to do an SAP implementation and they roll out an old E15K with 72procs. Beggers belief. Yes it will run the required number of users, but the end user experience will be a system running dog slow.
I've had customers who have migrated from 3.0ghz Xeons based blades to old Sun kit and yes the customers were not happy.
Also the latest Niagra chips are great for multithreaded apps were single thread performance is not an issue (serving web pages, java,etc) but your standard ERP system is mainly dependent on number of CPUs (multi-process) and CPU speed.
Too darn expensive
Absolutely fantastic servers, but there are cheaper alternatives. Anymore, the only time we use SUN is for our 'private' networks where we need extra security.
Lovely kit, but..
We used to be a Sun spot, so to speak, but one day some nutty Scandinavian called Linus or something wrote a little program called Linux....
Since then, our Sparcs have stopped coruscating, and we look at a long, long line of Poweredge boxes....
More simply put, Linux gave us the stability and familiarity of Unix, but without the $$$$ maintenance that Sun were charging, we could buy at least 2 Dell boxes for the cost of one Sun box, so we had redundancy.
Sure, it's not as elegant, efficient or pretty as the sun boxes, but we get equivalent uptime at a much reduced cost.
Since this decision Sun have made changes that may have kept them as our Unix supplier but it was too little, too late, and I feel that a lot of other folks have made the same choice - I seem to remember that back in the late 90s Sparcs and Solaris powered the internet - late in the naughties, it's Apache on Linux on anything you want...
dont count on FJ being the white horse.
I happen to work within the delightful company known as Fujitsu. It's often unaffectionately but passionately known as Fu-shit-su by many.
Longevity, longevity, longevity
Do any other manufacturers *guarantee* that a piece of code compiled 10 years ago will run now? It is the stability of the SPARC architecture and Solaris ABI which are the reasons people have bought Sun, and should do so now and in the future.
Companies don't buy hardware, they deploy applications, and nowadays a pretty small proportion of the total cost is kit. If deploying onto SPARC/Solaris means that your application will last 7 years rather than 5, you will save shed loads of cash.
SUN Failed in their marketing
SUN's problems started with the dot com burst and linux gaining popularity . Though Jonathan is trying to bring the ship back position he has to deal with legacy issues and a management who has no clue how to help him .
Though he knows his problem is marketing he has not done anything about it so number of years now (why the hell he keeps so many women in top) and that is going to hurt the company badly . It is time he hires some good sales guys from Microsoft/Cisco/Oracle who can sell shit load of stuff to customer with their good marketing guys . And Borrow some tips from guys like Cisco "upsell/weeklycommit/lockin".
SUN should reduce their R &D budget for sometime and use that for Sales , marketing and hiring some fresh energy . (Ask Gates he may be able to help with his free time)
Change some basic stuff - Stop SUN people using Windows (fire them all ) , stop advertising " migrating to new SAP" even if u do dont tell others about it (does MS ever tells u they use Linux servers) . Fire all those came from SAP & IBM and hire DELL & Cisco guys .
Change your focus on key products - Sunray - a good product SUN have no clue how to sell it and for that matter to price it . That should be your loss leader . If I am the marketing guy i will give a sunray free with every SUN server and every SUN monitor . (Or simply build the electronics inside the monitor and sell it for cost ) and people will buy Solaris and SUN servers for sure . What is the Point of advertising a Intel Server and use that energy on a CMT / JES .
What SUN needs is a fresh thinking in their marketing approach .. and sure things will fall in place .
Finally at this difficult times the size shouldn't bother SUN rather survival is key .
Still companies like riverbed / Bluecoat is in No1 position and give a good fight to Cisco WAAS and guys like Tandberg / polycom dont give up against Cisco Unified comm's , just because Cisco has a huge market CAP.
SUN is unique and Good from the comments above and SUN is important to this earth (sure we need few good guys in earth) .
SUN shouldnt get worried even if they have a bad fourth quater ( that may be because i didnt book that order i promised) but that will come in the next quater . No one is going to take away SUN's business ... trouble is times are bad .
For all the mournful Sunshiners posting here....
In best Nelson Muntz styleee - Ha ha!
Hey, maybe NetApp have $7bn in change laying around! Now that really would be funny!
The level of hatred of the Sun executives is just WOW!!!
Why can't people just get along? Jonathan has a pony-tail. He must be good!
Re: For all the mounful Sunshiners posting here....
Matt, you seem to take this all very personally.
I'm not sure what is sadder, the possibility of losing a company like Sun or your one man, misguided crusade to bring down a multi national company using a tech news site message board.