Apple Introduces Xserve Rackmount Servers

2nd Post! writes "MacCentral is reporting the announcement of 1U Apple rackmount hardware. The Xserve, despite its cheesy name, seems quite powerful: dual G4/1GHz with 4MB DDR L3 cache, up to 2GB DDR (yes!) SDRAM, 4 ATA drive bays (up to 480GB), 2 Gb Ethernet ports, 2 64/66 PCI slots (one of which may be taken up by one Gb Ethernet card), and, of course, FireWire. Pricing starts at $2,999 for a single 60GB disk and 256MB RAM." Yahoo! has posted the press release; Doc Searls is writing about Jobs' speech. Update: 05/14 18:14 GMT by M : Apple's page about the Xserve is now live.

Comments

Oracle 9i Too!

[rgraham]

This wasn't mentioned in the press release but seems like a pretty big deal and come from the MacCentral coverage: "Introduces Mike Rocha, senior vice president, Platform Tech, Oracle: Oracle 9i on OS X -- we very excited about this hardware. Oracle is about low-cost clustering. Future releases will be on-time, synchronous. When we use UNIX native support, native APIs, optimized for this hardware, we can synchronize our releases so that our customers can have unified database versions across different hardware platforms. "

Re:I don't get it - for me, quality

[victim]

For me, having a quality hardware product with a reasonably secure OS that just works on it is the attraction.

The last batch of 6 1U x86 rackmount servers I bought from one of the largest PC manufacturers came with misprogrammed APICs that made them unable to run Linux without spending several days on hackery to get them going. The PCI slots are still useless, they can't deliver interrupts, but the rest of the machine works. I shuttled machines around so they don't need their PCI slots. (This machine was not purchased with Windows, it was a no-OS machine.) Two of these machines have failed in the 6 months that I have owned them.

The previous batch of 2U servers I purchased had a whiz-bang scsi controller that displayed a linux allergy and took me weeks of trying pre-release patches and waiting to get a linux version that worked acceptably. I still have to build custom kernels for these machines when I upgrade.

The biggest problem I have purchasing PC hardware is there is no good way to tell what is "server grade" and what has cheaped out components in the power supply or capacitors that will cause their MTBF to suffer. The extreme price pressure always tempts the manufacturers to cut corners.

So, the attactions...

• Apple (with a couple of stunning counterexamples (AppleIII, first Airports, some monitors)) was an outstanding reputation for making high grade hardware.
  
• The OS is going to work correctly on the hardware.
  
• If I like the machine, I will be able to order more identical machines 6 months later.
  
• I will not be rolling the dice to see if my OS will run on the new hardware.
  
• The firewire ports will work. Even with two processes hitting the same disk at the same time.

Ok, they cost about 30% more than the servers I have been buying (and certainly outperforms them, but that is irrelevant, my servers are low cpu users). I'll take that. It vanishes in the unbilled hours dealing with mystery hardware and having to buy a bunch of spare machines to count on being able to replace a machine when needed.

Top500 time?

[dbirchall]

Yeah, those are pretty nice specs. It gets a little more interesting when you take that theoretical peak performance of 630 GFlops for a rack of these babies and look at the most recent Top500 list.

A lot of us snickered when Apple pitched the G4 as a "supercomputer" (using the technical export definition), but if folks like Genentech build racks of these, clustered, and land in the top 10% of the Top500 list, Steve and company will be the ones laughing.

Let's see... the *bottom* of the Top500 list is currently a 116-CPU Cray T3E 1200, with a theoretical peak of about 139 GFlops... you'd only need enough Xserves to fill 1/4 of a rack to come up with that kind of power.

Okay, okay, I guess I want some too.