Sun Joins Apple in the Intel Camp for x86 Chips

An anonymous reader writes "Don't worry, SPARC isn't being replaced by Itanic. However, Sun will start using Intel Xeon CPU's in their X86 servers. Further evidence that Intel's Core microarchitecture is winning back a lot of the business that AMD won with Opteron." More coverage at CNN Money and the International Herald Tribune.

Comments

This is a surprise

[Tim C]

I was at JavaUk06 last year, and in his keynote speech (one of) the Marketing VPs spent quite a lot of time extolling the virtues of their new line of SunFire servers, paying particular attention to their power:performance ratio compared to similar Xeon-based servers. Listening to him then, you'd have thought that Opterons were the best thing since sliced bread. Yes, I realise that his job is to push their current and up-coming products and solutions, but the main thrust of his talk was "Opteron-powered SunFire servers use far less power than those crappy, power-hungry Xeon servers".

Re:This is a surprise

[be-fan]

If this was last year, he was likely talking about the older (P4-based) Xeon, not the new (Core 2-based) Xeon.

Re:This is a surprise

[Anonymous Coward]

"you'd have thought that Opterons were the best thing since sliced bread."

In fact it was at the time when it was competing against Xeon's based on the P4 core. Things have changed since the new Core2 based Xeon's have come on the market.

Re:So, ahhhh...

[TheThiefMaster]

No, they're using Intel chips in their line of servers that previously used AMD chips. For the pro-AMD slashdotters, this is "a very bad thing"(tm).

sun.com

[Daemonstar]

Of all the links posted in the summary, there's no link to the webcast on Sun's site about the story (01/22/07 @ 10:00 PST, Realplayer 10 required). :P

Sun is not abandoning AMD. Sun is adding Intel.

[gp310ad]

Which makes sense. When there are two competitive players whose product features and performance keep passing each other, why not give the customer a choice and at the same time exploit that competition to improve ones own position...

Re:Sun needs this

[nwhitehorn]

Sun is moving ahead with their SPARC servers, and just taped out a successor to the Niagara. If you'd read the article, you'd know they are replacing their (quite excellent) AMD servers with Intel ones, not SPARC with anything. Sun has quite happily been selling both architectures for some time now.

Re:Sun needs this

[Anonymous Coward]

I hope this marks the beginning of Sun moving entirely to Intel/x86 based chips, this way Sun can focus on their other ailing businesses. Sun (just like Mac) is not big enough to keep up with AMD and Intel on chip performance, so why spend Millions/Billions trying?

Don't count on it just yet. As a former Solaris jockey and proud owner of a Core 2 Duo E6700 on my desktop (it screams!), I think this is a smart move at the lower end of their server market. As far as the high end goes, I don't see Intel's stuff scaling (n > 16 CPUs) as smoothly as the traditional RISC brigade at the moment. Anybody with actual enterprise experience please feel free to correct me if I'm wrong.

Re:I'm sceptical

[TheRaven64]

I say Good for Intel, the Core CPU is a good one. But, if you look at everything Intel has been doing with their CPU line lately, you'll see that they are generally copying AMD in a lot of places, starting with EM64T (aka AMD64.) Copying from AMD? Not really. Intel had dual-core before AMD (although you could say they copied this from IBM). They had SMT (HyperThreading) before AMD, although seem to have abandoned it in their recent chips (again, IBM had this first too). x86-64 extensions? Sure, but they are some minor tweaks to the ISA. SSE4 is a similar sized change, and they didn't copy this from AMD (although AMD are now implementing SSE in their chips...). What about dynamic shared caches? Nope, Intel got this one before AMD too (IBM, though, had it before either). 65nm process technology? Nope, Intel are transitioning away from this just as AMD are moving to it.

The only place AMD are really ahead at the moment is in processor interconnect, and in most situations interconnect throughput is not a bottleneck for Intel (latency, on the other hand, is a big problem for everyone, although Intel have better pre-caching technology than AMD at the moment, making it less of a problem for them). Oh, AMD also have a slight advantage in virtualisation (their extensions are a bit better than Intel's), but they are both so far behind the likes of IBM and Sun in this area that it's not even funny.

Re:Oh Gosh! Sun 386 all over again?

[h4ck7h3p14n37]

My condolences. I've managed Solaris systems for years, almost entirely SPARCs, but the new place I'm working for buys the cheap x86 junk. Now the x86 systems that Sun ships aren't bad despite some issues with the nVidia RAID chip, Intel Gigabit ethernet nics and the damn x86 BIOS, but running it on a generic 1U x86 server is a joke compared to SPARC. You've got no lights out management support (fortunately some of Sun's AMD systems come with a dedicated service processor for this task, but it requires additional setup), the system boards typically support SATA disks and not SAS or SCSI, which in my experience are much more reliable and provide better feedback to the operator when they're starting to fail, and Solaris 10 currently has a major reliability problem on the x86 architecture.

If one of your Solaris 10, x86 systems goes down improperly, it's most likely _not_ going to come up without human intervention. I still don't understand the whole process, but the second stage boot loader reads a boot archive into memory; if there have been changes made to the root filesystem since the archive was last updated, the system will not boot. You'll have to fsck the disks manually, then clear the system/boot-archive service, then update it with "bootadm update-archive" and reboot. WTF? One simply cannot count on the x86 systems to come back up.

Give me a good 'ol SPARC based system with a boot ROM anyday; the x86 architecture is just plain fscked.

Re:Sun needs this

[TheRaven64]

Depends on the workload. For floating point, the T1 is pretty dire (the T2 is better). The thing it does really well is highly concurrent workloads. AMD have a faster interconnect than Intel at the moment, but they are both crippled by one thing: the speed of light (painfully slow). The amount of work that a modern processor could do while waiting for a cache miss is enormous. Intel currently have some very clever logic predicting and pre-fetching cache misses, but they still suffer from them. Sun, however, cheat. Each core on a T1 has 4 contexts (8 on a T2), so when you get a cache miss it just switches over to the next thread instantly and carries on executing. The T1 core will only sit doing nothing if all four cores are in the middle of a cache miss, while an Intel or AMD CPU will do this for a few hundred cycles on every single cache miss.

For desktop workloads, this isn't such a great thing; most current-generation desktop applications do all their work in one thread, so if that thread has a cache miss you still end up doing a load of waiting because the other threads are not using the CPU much. On a server with a few hundred (or thousand) concurrent users, however, there are always threads waiting to do something, so you can get a phenomenal amount of throughput from this. With the growth in web applications, I expect Sun to do very well.