Softpanorama (slightly skeptical) Open Source Software Educational Society

May the source be with you, but remember the KISS principle ;-) Softpanorama Search

Solaris History

Sun's logo with interleaved copies of the word sun, was designed by Stanford University professor Vaughan Pratt. The initial version of the logo with sides oriented horizontally/vertically was later changed to the box appearing to stand on one corner.

Sun Microsystems - Wikipedia, the free encyclopedia

Sun originally used the Motorola 68000 CPU family for the Sun-1 through Sun-3 computer series. The Sun-1 employed a 68000 CPU, the Sun-2 series, a 68010. The Sun-3 series was based on the 68020, with the later Sun-3x variant using the 68030.

Starting with the Sun-4 line, the company used its own processor architecture, SPARC, a 32-bit RISC architecture which was later to become the IEEE 1754 standard for microprocessors. A 64-bit extension of the SPARC architecture (SPARC V9) was later introduced.

Sun has implemented multiple high-end generations of the SPARC architecture, including SPARC, SuperSPARC, UltraSPARC-I, UltraSPARC-II, UltraSPARC-III, and currently UltraSPARC-IV. Sun has developed several generations of workstations and servers, including the SPARCstation series, Sun Ultra series the Ultra Enterprise (later, simply "Enterprise") servers, the Sun Blade workstations and the Sun Fire servers. Sun also has a second line of lower cost processors meant for low-end systems which included the MicroSPARC-I, MicroSPARC-II, UltraSPARC-IIe, UltraSPARC-IIi, and UltraSPARC-IIIi.

Sun has had a difficult time keeping up with its competitors' processors' clock speed and computing power, but its customer base has been fairly loyal due to the popularity, and legendary stability, of its SunOS (and later Solaris) versions of Unix.

For the first decade of Sun's history, the company was predominantly a vendor of technical workstations, competing successfully as a low-cost vendor during the Workstation Wars of the 1980s.

For a short period in the mid-1980s, 51% of Sun stock was held by AT&T as a partner in their computer business AT&T Computer Systems. UNIX System V Release 4 was jointly developed by AT&T and Sun, who named their version Solaris 2. The AT&T partnership later fell apart when the rival group OSF (Open Software Foundation) appeared. See UNIX wars.

For a short period in the late 1980s, they sold a hybrid Intel 80386-based machine, the Sun386i. An x86 port of Solaris for PC compatibles was introduced in 1993. Currently, Sun is again selling x86 and AMD64 hardware and has introduced a 64-bit version of Solaris for AMD64 systems.

In the mid-1990s, Sun acquired Diba and Cobalt Networks with the aim of building network appliances (single function computers meant for consumers). Sun also marketed a network computer (diskless workstation, as popularized by Oracle Corporation CEO Larry Ellison). None of these business initiatives were particularly successful.

In the late-1990s, as Sun's workstations were lagging in performance when compared to that of their competitors and especially to Wintel Personal Computers, the company successfully transformed itself to a vendor of large-scale Symmetric multiprocessing servers. This transition was enabled by technology that was acquired from Silicon Graphics and Cray Research. The Cray CS6400 server line was transformed into the very successful Sun Enterprise 10000 large-scale servers. Driven by the increased prominence of web-serving database-searching applications, blade servers (high density rack-mounted systems) were also emphasized.

The Sun 1 was shipped with Unisoft V7 UNIX. Later Bill Joy, the key figure of the Berkeley Software Distribution (BSD) and one of four founders of Sun produced a customized 4.1BSD UNIX called SunOS as an operating system for its workstations.  Up through version 4.1.x (Solaris 1.x), SunOS remained a heavily BSD-influenced Unix implementation.

In the late '80s, Sun entered into a partnership with AT&T, which was then developing the other major Unix flavor, System V. The result was System V release 4 (SVR4), which incorporated BSD as well as SunOS extensions (e.g., NFS). Subsequently, with its version 5.x (Solaris 2.x) releases, SunOS shifted from its BSD origins to SVR4.

To confuse things Sun Solaris is sometimes called SunOS 5, while old version of SunOS up to 4 are referred as for Solaris 1). To further confuse the naming scheme, Sun now refers to Solaris by just its point release (e.g., Solaris 7, 8, or 9 instead of 2.7, 2.8, or 2.9).

For more information about SunOS and Solaris, including FAQs, white papers, upgrade, and purchasing information try Google. 

History of Sun and Solaris from 1991 till 2001 is partially reflected in Sun under the Linux siege

You can also consult the following newsgroups:

• comp.unix.solaris
• alt.solaris.x86 
• Those in the comp.sys.sun.* hierarchy

Sun Hardware Timeline

Chronology of Workstation Computers

1982

February

• Scott McNealy, Bill Joy, Andreas Bechtolsheim, and Vinod Khosla found Sun Microsystems. "SUN" originally stood for Stanford University Network. [47] [110.149,152] [217.163]

May

• Sun Microsystems begins shipping the Sun 1 workstation computer. [110.152]

1984

February 14

• Scott McNealy is appointed president and chief operating officer of Sun Microsystems. [110.153] [218.D2]

April

• Silicon Graphics begins shipping its first 3-D graphics workstations. [28]

June

• Motorola introduces the 16 MHz 68020 processor, a 32-bit version of the 68000, in CMOS, with on-board cache. [1] [140] (1986 [20])

(month unknown)

• MIPS Computer Systems is founded, and begins developing its RISC architecture. [29]
• Sun Microsystems co-founder Vinod Khosla resigns. [110.153]
• Silicon Graphics introduces its first workstation, IRIS 1400. [221.61]

 

Sun Hardware

Sun 1's

These are the large black desktop boxes with 17" monitors. Used the original Stanford-designed video board. Uses a parallel microswitch keyboard and parallel mouse.

1/100

Used design similar to original SUN (Stanford University Network) CPU, version 1.5 CPU could take larger RAMS. Pre-dates Sun's 4.2 port (ran Unisoft V7) (68010 CPU instead of SUN's 68000) 10Mhz.

1/100u

"Brain transplant" for 100 series. Replaces CPU and memory card with first-generation Sun2 CPU and memory boards so original customers could run SunOS V1. (Still has parallel kb/mouse intf so old kbds would work.)

1/170

Rack-mounted server. Slightly different chassis design than 2/170's

---

Sun 2's

2/120

Multibus-based 68010 10Mhz. First machines that had desk-side chassis Serial Microswitch keyboard, Mouse Systems Optical mouse. 8Mb memory max. Cards are CPU, 1 or 4 meg memory board, ethernet board, SCSI board, 640 * 480 color board, monochrome video board, SMD controller, tape controller, 16 port serial mux (ALM-1)

Two variants of video board, one generated TTL-level video, on ECL. Later video boards ("2prime") could generate either levels. Early 19" mono monitors (philips or moniterm) could be switched as well.

2/170

VME Sun2 style CPU 2 slot chassis. Optional SCSI board (model name is SCSI-2; 2'nd SCSI design.. first was for 2/1xx's) sat on mem expansion board in 2nd slot. CPU board had 1,2,or 4 megs mem, 10Mhz 68010 CPU, ethernet, two serial ports. Memory expansion boards are 1,2 or 4 megs as well. The (type-2) keyboard and mouse attached via an adapter that accepted 2 modular plugs and attached to the DB15 port.

2/160

First machine to use 12 slot desk-side VME chassis. Many have CPU upgrades to 3/160's. Had 4 fan cooling tray instead of 6 in later machines, thus cooling problems with lots of cards. Also only had 4 P2 memory connectors bussed instead of 6.

SunOS 4.0.3 was the last release with Sun2 support.

2/1xx's with a monochrome display can only have 7megs max, since the frame buffer appears in the 8th meg

Sun 3's

3/160

First 68020 based Sun machine. Uses "Carrera" CPU, which is used in lots of other Sun3 variants. 4Mb on-board memory. Sun's mem expansion goes on 4 Meg memory expansion boards; third parties had up to 32 megs on one card. SCSI was optional. One variant of the memory card held the 6u VME SCSI board, other version sat in slot7 of the backplane and ran the SCSI out the back of the backplane to the internal disc/tape. CPU has 2 serial, ethernet, kbd ports.

 

I, Cringely . September 9, 1999 - Terminal Condition PBS Why Sun's Aggressive New Workstations Are Really Just a Blast From the PastBy Robert X. Cringely

Each week, I have to decide a topic for this column. The problem is not finding a topic, but choosing one from the many obvious candidates. A few times, I've tried to cover more than one topic, but there is a firm nerd contingent among my readers who think we have a contract allowing only a single topic per week. I am not here to argue, so they win. But this week I am torn, since there are obvious developments in the news as well as the 30th anniversary of the first Arpanet node coming to life. Or I could even try to explain why last Week, I thought former Secretary of Defense Robert McNamara's first name was "George." The answer to this last mystery is easy — dementia. As for the Arpanet birthday, having done three hours of Internet history for television and written several columns on the subject already, this time I'll just congratulate all concerned with those events back at UCLA: It has been quite a ride.

That first Arpanet Interface Message Processor (IMP) was built from a Honeywell computer used by the military. In the early 1980s, the Internet building block of choice came to be logic boards for Sun workstations. The first Cisco routers, for example, were built from Sun logic boards designed by Andy Bechtolscheim when he was a graduate student at Stanford. So too, the first Silicon Graphics workstations were Sun workstations with extra 3D capability added-in by Jim Clark. All of these companies were founded in the same building and all are still on the scene, but this week, I'd say Sun has been making the most news, though in a disturbingly regressive manner.

Last week, I wrote about Sun's acquisition of Star Office, and how this would put a virtually free office suite up against Microsoft for both PC- and server-based versions. Well, this week the other shoe dropped as Sun introduced its candidate workstation for the server-based version. It is a funny little box called the Sun Ray 1 Enterprise Appliance, into which you attach a keyboard, mouse and screen, then use an Ethernet connection to the world. The Sun Ray looks to be a successor to Sun's own unpopular JavaStation and the logical heir to the network computing crown. Or is it?

The Sun Ray is great from a configuration standpoint, since it requires no configuration at all. You couldn't configure it if you wanted to. If the box breaks, you replace it with another. Plug it into power and Ethernet, and it is ready to go. This is all marketing talk here, but reading it I came to have an unsettling feeling. Then it came to me. The Sun Ray, for all its high design and ease of use, is not a computer at all or even a computing device. It is a computer terminal. Sun's answer to Microsoft is to take corporate America back to a souped-up version of 1970's minicomputing.

The only application that runs in the Sun Ray is whatever paints the screen and accepts keyboard and mouse input. That's a computer terminal where I come from. Presumably, there is a TCP/IP stack and something like an X-Window server, though Sun does an excellent job of not telling us that. What's definitely NOT happening in the box is anything like Java, which Sun has finally figured out isn't up to the task. Instead, all the real computing is done back on a hefty Sun server and only screen rendering happens in the Sun Ray.

There is another outfit called Network Computing Devices that makes boxes like this, which it calls X-terminals. NCD was founded by Bill Carrico and Judy Estrin, a husband and wife team who also founded Bridge Communications (later part of 3Com) and Precept Software (later part of Cisco). Judy is now the Chief Technical Officer at Cisco, which fits perfectly into my theory that there are really only 25 people in the computer business. They just keep changing jobs. I remember visiting Bill and Judy late in their tenure at NCD, a time that wasn't particularly happy since X-terminals were being rapidly replaced with cheap PCs running X-server software. In the world of X, what we would normally call a "client" — that part of the application that runs on the workstation rather than on that big box in the computer room — is called a "server." Go figure.

The wonderful thing about an X-terminal is that it does an end-run around user ego. NCD boxes were all connected through Ethernet to a Sun server. How many X-terminals could a Sun server serve? Lots. Typically 25-50 terminals could be run by a single server that cost a lot more than a PC, but sure didn't cost 25-50 times as much. Still, that day I visited Bill and Judy, they saw the end coming. Why? Because PCs were cheaper than X-terminals and they could run local applications, too. With PCs even cheaper today, what has changed to make Bill and Judy wrong and Sun happy to enter this new business?

Well NCD, which is still very much in business, didn't sell servers, and Sun does, so that's an enormous difference. Sun makes its dough on this deal not from the Sun Rays or from Star Office, but from the big iron it sells to support both. And in the last few years, the world of corporate computing has come to fixate on Total Cost of Ownership (TCO), which includes everything from the cost of training users to replacing busted boxes. In the grand scheme of TCO, the original purchase price of a PC is almost insignificant, dwarfed by the human cost of setting-up and shifting and training, etc., all of which are minimized by the Sun Ray/Star Office combo. On a TCO basis, the Sun Ray is damned cheap, and with Sun offering leases at under $10 per month, it is a good deal for many businesses.

But don't expect to run your Sun Ray at home because its sparkling performance has more than anything to do with that 100 megabits-per-second Ethernet connection. Running over a 56K modem won't work at all. And don't even think of using a Sun Ray unless you want at least 25 of them, because it's only at those scales that the costs begin to come into line. What this means, then, is that the Sun Ray is far from a Microsoft killer. Rather, it is a Microsoft annoyance. But for Sun, it is still a very good business.

The nerds will say this is obvious and that I'm again wasting their time, but most of the people who read this column aren't nerds. They'll say, "Now I get it."

Solaris Timeline

1979 - Bill Joy introduces "Berkeley enhancements" as BSD 4.1.

1982  - The company was incorporated in 1982 and 1986. Founders include Vinod Khosla, Scott McNealy, Bill Joy and Andy Bechtolsheim. Of these men, only McNealy and Bechtolsheim remain with Sun.Sun Microsystems is founded by Vinod Khosla, Andy Bechtolsheim, and Scott McNealy with $4 million in venture capital with four emplyees.

1983  - Sun Microsystems introduces SunOS.

1984 - About 100,000 UNIX sites exist worldwide. Sun now has 400 employees and $39 million in annual sales.

1986 - Sun went public in 1986

1988  - AT&T and Sun start work on SVR4, a unified version of UNIX.

1988 - OSF and UI are formed.

1989  - AT&T releases System V, release 4.

1990  - OSF releases OSF/1.

1992 - Sun introduces Solaris, which is based on System V, release 4. SunOS, which is based on BSDF UNIX, will be phased out. Sun now has more than 12,500 employees and more than $3.5 billion in sales.

1993 - Novell buys UNIX from AT&T.

1994 - Solaris 2.4 is available.

1995

• Santa Cruz Operation buys UNIXware from Novell. SCO and HP announce a relationship to develop a 64-bit version of UNIX. Solaris 2.5 is available -- the  first stable version of Solaris 2

• Bechtolsheim, 48, left Sun in 1995 to start Granite Systems, which built 1-gigabit-per-second networking technology and which Cisco acquired in 1996.

1997 - Solaris 2.6 is available. It soon became the most popular version of Solaris for the next three years

1998 - Solaris 7 is available.  -- This was not very successful version. Few moved from 2.6 to 2.7

2000 - Solaris 8 is available.  Became a huge success. Most move from 2.6 directly to Solaris 8

2003  September Bill Joy, Sun Microsystems chief scientist and co-founder, is leaving the company, moving on to "different challenges". No, he's not saying yet what those different challenges are. Greg Papadopoulos, CTO, will take over Joy's responsibilities. See09-09-03 - SUN MICROSYSTEMS ANNOUNCES CHIEF SCIENTIST BILL JOY TO LEAVE COMPANY

2004 February Nine years after leaving the server maker he co-founded in 1982, Andy Bechtolsheim is returning to Sun Microsystems.

Old News ;-)

[Apr 24, 2009] The downfall of Sun Microsystems by Jon Brodkin

04/24/2009  | Network World

Dot-com bust, failure to embrace x86 processors ended Sun’s life as an independent company, analysts say

Oracle's surprising $7.4 billion deal to purchase Sun this week gives Larry Ellison and crew a big stake in the hardware market as well as control over Java and other well-known open source technologies. But it also spells the end of an independent Sun Microsystems, one of Silicon Valley's most prominent companies.

How did it all come to this for Sun, often regarded as one of IT's great innovators during its 27-year lifespan? The dot-com crash at the start of this decade is frequently cited as the beginning of the end for Sun, and for good reason. Acquisition missteps and a failure to monetize key products such as Java also hastened Sun's descent.

"The dot-com bust hurt everybody but it's arguable that Sun was hurt most of all because it had profited so much in the run up to the boom in the first place, and hadn't grown its business out as deeply as IBM and some others had," says Pund-IT analyst Charles King.

Sun's Sparc servers with the Solaris operating system were snatched up by dot-com start-ups because of their stability and flexibility in deploying various applications at affordable prices, King says.

"In the months following the bust, there was a huge amount of Sun product that was out on the street and it precluded the need for people to upgrade or purchase new equipment," King says.

Sun prized its Sparc architecture so much that it missed the industry-wide transition to x86 processors, analysts say. Sun actually did sell x86-based systems in the 1980s, but concentrated its efforts on Sparc for most of the 90s. In King's view, Sun treated x86 systems as nice toys, but not platforms that could be used to power a serious corporate data center. Sun did increase its presence in the x86 market in the years following the dot-com bust with AMD- and Intel-based servers, but it seems to have been too little, too late.

The biggest reason for Sun's downfall is "the inability to recognize the x86 open architecture, as opposed to what they were selling with the Sparc processors," says Enterprise Strategy Group analyst Brian Babineau.

Babineau also faults Sun for pursuing a "non-capitalistic strategy" by emphasizing open source, yet failing to monetize key products such as Java.

King and Babineau both point to failed acquisitions. King notes Sun's $2 billion purchase of Cobalt Networks, a server appliance vendor that was gobbled up by Sun in 2000 but never produced any real dividends for its owner.

Sun has attempted to compete in many different hardware and software markets, but is too often in third or fourth place, Babineau says. Sun bought MySQL for $1 billion in 2008, for example, challenging the database market where Oracle was already king. Sun also executed poorly in the storage market after purchasing the vendor StorageTek for $4.1 billion in 2005, Babineau says.

"There was just mismanagement," Babineau says. "They purchased so many different things over the years. It was panic and frantic at the end."

Following the dot-com crash, Sun's profits took an immediate dive. After reporting net income of $1.85 billion in fiscal 2000, that number was halved to $927 million in 2001. Sun lost $628 million in fiscal 2002 and a whopping $2.4 billion in fiscal 2003. It returned to profitability in fiscal 2007, but ultimately the company reported net losses in three of the four most recent quarters, and the sharks started circling, in a manner of speaking. IBM offered $7 billion to buy Sun, only to be rebuffed. Several analysts doubted that Sun could find another buyer after rejecting IBM, but then Oracle came calling.

One reason Sun couldn't go on in its present form is that the company had a core group of loyal customers but wasn't able to win many new accounts, King says. And for many years, when Sun's customers wanted a reliable x86 platform they had to turn to Sun's competitors.

Related Content

"The history of the Valley is littered with the dried husks of companies that had great technology but didn't understand the dynamics of the commercial market they were trying to compete in," King says.

That's not to say Oracle won't be able to gain success with Sun's technology. While Sun has failed to maintain profitability, the company did pull in more than $3 billion in revenue in the most recent quarter.

Oracle is touting Java and Solaris as two key software assets that will help Oracle and Sun turn a larger profit than they could separately. Oracle, which is expected to significantly reduce Sun's expenses, predicted that Sun will bring $1.5 billion in operating profit in its first year as part of the combined company.  

"Java is one of the computer industry's best-known brands and most widely deployed technologies, and it is the most important software Oracle has ever acquired," Oracle said in a statement announcing the acquisition. "The Sun Solaris operating system is the leading platform for the Oracle database, Oracle's largest business, and has been for a long time. With the acquisition of Sun, Oracle can optimize the Oracle database for some of the unique, high-end features of Solaris."

With Oracle seemingly most excited about Sun's software platforms, Babineau speculates that Oracle might ultimately sell off the hardware business. Other analysts, such as Forrester Research's James Kobielus, say Oracle should leverage its new hardware capabilities with data warehousing appliances that integrate MySQL and other Oracle databases into Sun servers.

On the whole, Oracle's announcement of the purchase was "remarkably devoid of detail," King says, so it's tough to say what the combined company will look like one or two years from now. Oracle and Sun had such tight partnerships already that dramatic changes may be the exception rather than the rule, he says.

"Frankly Oracle and Sun have worked very closely for the better part of two decades and I don't really see what the companies will be able to do as a single organization that they haven't already done as close strategic innovative partners," King says.

Selected Comments

Synergy and Proven Management By Vic W. on Fri, 04/24/2009 - 8:02am

It's clear that Oracle is driven to succeed, their history and leadership offer continuous evidence of that. Looking from the outside as a competitor: They are a very well managed company, taking as much of the revenue of any contract or sale as they possibly can. In business, a true predator.

Sun has been a engineering company first and foremost, innovation and engineering brilliance are what make Sun standout among the surviving computer companies. (To be fair, Apple is much the same -- but Apple is in a very different market and Apple is very well managed these days). Sun has not been well managed for a long time.

Sun had so much engineering and product DNA that Oracle can easily manage these pieces for many years to come, reaping profit off of what Sun itself failed to monetize to the scope of where Oracle will surely take it.

With this Sun DNA, Oracle will be competing directly with IBM and HP. Of these companies, Oracle has the predatory business edge. IBM and HP seem hell bent on supporting their bottom line with excessive services revenue -- which can easily fall under the cost benefits of automation and better composed product suites.

The Sun DNA will be seen in Oracle's products for at least the next decade. This acquisition is a brilliant move by a well run company that not only wants to grow but that also has huge ambitions and

[Jul 3, 2007] Multiple Threads Phil Harman's Weblog A Brief History Of Solaris

A week ago I was presenting A Brief History Of Solaris at the Sun HPC Consortium in Dresden. My slideware is pretty minimalist (audiences generally don't respond well to extended lists of bullet points), but it should give you a flavour of my presentation style and content. For more, see Josh Simon's writeup.

My main point is that although Solaris is a good place to be because it has a consistent track record of innovation (e.g. ONC, mmap, dynamic linking, audaciously scalable SMP, threads, doors, 64-bit, containers, large memory support, zones, ZFS, DTrace, ...), the clincher is that these innovations meet in a robust package with long term compatibility and support.

Linus may kid himself that ZFS is all Solaris has to offer, but the Linux community has been sincerely flattering Sun for years with its imitation and use of so many Solaris technologies. Yes, there is potential for this to work both ways, but until now the traffic has been mostly a one way street.

As a colleague recently pointed out it is worth considering questions like "what would Solaris be without the Linux interfaces it has adopted?" and "what would Linux be without the interfaces it has adopted from Sun?" (e.g. NFS, NIS, PAM, nsswitch.conf, ld.so.1, LD_*, /proc, doors, kernel slab allocator, ...). Wow, isn't sharing cool!

Solaris: often imitated, seldom bettered.

[Jul 3, 2007] The Navel of Narcissus

Phil Harman from Sun's Solaris group gave an informative and amusing talk at the HPC Consortium meeting in Dresden this week titled, "A Brief History of Solaris." I'm hoping the full talk will be posted on the Consortium site at some point.

Phil began his history of Solaris by reminding us of some of the "prehistoric" innovations in SunOS. For example, who but Sun was doing open network computing back in the 1980s with innovations like NFS, NIS, the automounter, XDR, and RPC? How about the STREAMS abstraction? mmap? ld.so?

He then moved to innovations done by Sun "within living memory." His list included loadable, configurable kernels; dynamic system domains; /proc; truss; the p-tools; and /etc/nsswitch.conf. Not to mention "audacious" SMP scalability, and a compatible 32/64 bit transition strategy that maintained binary investments through our transition to 64-bit computing. Oh yes, and there was that Java thing as well...

Innovations done "just yesterday" included Hierarchical Lgroup Support (HLS), Multiple Page Size Support (MPSS), containers, Service Management Facility (SMF), zones, BrandZ, ZFS, and DTrace.

He finished with some comments on ZFS, which he motivated with the graphic I've placed at the top of this blog post. It illustrates the problems of single-bit errors. In this case, a printer was fined by the King of England for what amounted to a life's wages for making this error in a 1631 edition of the King James bible (known as the Wicked Bible). "Got checksums?", asks Phil as he noted that ZFS protects the datapath all the way from the rotating rust (the disk) to memory.

Does the "I" in RAID mean "Inexpensive" or "Independent"? The former is correct, so why do some in our industry prefer the "independent" interpretation? Phil explained why during his talk and also in this blog entry.

[Dec 26, 2006] McNealy's cold feet and other tales of Sun Tech News on

Jan 12, 2006 (ZDNet) MOUNTAIN VIEW, Calif.--Sun Microsystems CEO Scott McNealy had to be wined and dined at a Silicon Valley McDonald's before he gave up his reluctance to help launch the workstation maker in 1982, according to one of many tales the company co-founders recounted on Wednesday.

McNealy joined Sun's other co-founders, Vinod Khosla, Andy Bechtolsheim and Bill Joy, at a panel discussion at the Computer History Museum here to reminisce about the server specialist's past and prognosticate about the future.

Khosla said the McDonald's meal took place just after he and McNealy met with venture capitalists and got Sun's first funding commitment. "We went out and sat in the parking lot. Scott said to me, 'I don't know if I really want to do it.' So I took him to an upscale dinner at McDonald's on Page Mill Road" in Palo Alto, Calif., he said, where he put the screws on McNealy to resign from his $40,000-a-year job at Onyx Systems.

"Vinod asked me, 'When are you quitting?'" McNealy recounted. When McNealy balked, Khosla countered, saying: "'You can't back out on me now. You're a founder.' "I said, 'Oh, OK.' It was that quick," McNealy said.

Khosla left Sun in 1986 to become a general partner at venture capitalist firm Kleiner Perkins Caufield & Byers, and Joy followed suit in 2005. Bechtolsheim left Sun in 1995 to found gigabit Ethernet start-up Granite Systems, later acquired by Cisco Systems. But he rejoined the company in 2004, when Sun bought his next start-up, Kealia, to provide the foundation of its new Intel-based Macs).

"We got very close to having Apple use Sparc. That almost happened," Joy said.

In total, "there were six very, very close encounters" with Apple, he noted. That none of them worked out was a "personal disappointment" said Joy, who spent years as Sun's chief technology officer.

McNealy added that he went to Steve Jobs' house to try to hammer out the user interface agreement. The Apple co-founder and CEO was "sitting under a tree, reading 'How to Make a Nuclear Bomb,'" with bare feet and wearing jeans with holes torn in the knees, McNealy said. The interface work, though, "never went anywhere," he said.

Khosla also lavished praise on Jobs, who he said was a role model, along with Oracle CEO Larry Ellison and Intel's former CEO Andy Grove. Jobs is the kind of person "who passionately, religiously believes his own ideas. No matter what anybody else says, he's going to push them through," and that determination and self-confidence is in large part why he succeeds in doing so, Khosla said.

McNealy has praised Jobs on occasion, but he acknowledged on Wednesday that he doesn't have time to listen to his own iPod and forecasted doom for the popular digital music player. The right place to store music is on the network, where it can be accessed by many devices, he said, much like the right place to store voice mail is on a central server.

"Your iPod is like your home answering machine. It's a temporary thing," McNealy said. "It's going to be hard to sell a lot of iPods five years from now, when every cell phone is going to be able to automatically access your library wherever you are."

[Jun 14, 2005] Yuzo Watanabe A Brief History of Network Driver Development in Solaris

[Feb 11, 2004] InformationWeek Sun Microsystems Acquisition Brings Back Sun Microsystems Founder

Andy Bechtolsheim, who left Sun in 1995 to pursue other business opportunities, currently leads Kealia Inc., which develops advanced server technologies. Sun's acquisition of Kealia is expected to close later this year, the companies said Tuesday. Sun spokeswoman May Petry declined to disclose terms of the deal.

Recommended Links

---

In case of broken links please try to use Google search. If you find the page please notify us about new location

Sun Microsystems - Wikipedia, the free encyclopedia

history_of_solaris

UNIX History

Solaris_History_6per

Sun 3/3x Archive warmed our hearts and made us remember the "good ole days". Get all the info for these Motorola 68020- and 68030-based systems. FAQs, part numbers, benchmarks, error codes, patches, and more.

OpenBoot Questions & Answers [1995] is good for diagnosing hardware problems, reading device trees, and understanding Sun's OpenBoot (boot prom) in general.
 

The Sun Voyager FAQ has everything you want to know about the hardware and software of this machine.
 

Using SPARCPrinter with Ghostscript covers the basics of getting this working on Solaris 2.6.

Version History Reference

SunOS & Solaris Version History