Finally found my E2180! Chip en route, overclock results to come

I’ve been doing sporadic checks on a few major vendors’ sites and on price comparison sites on when this mythical CPU may appear. As luck would have it, I finally found the CPU for sale at eWiz last night. Put in an order for the elusive thing, and hopefully I will have my hands on it sometime next week. For approx. $95 including shipping and CA tax, it wasn’t so bad a deal in my opinion.

Reports from many forums indicate that even the E2140 and E2160 are starting to show up with the M0 stepping. These things are supposed to be monster overclockers, with reports of them clocking just as well as their G0 siblings, providing the boards are capable. Now that I have my Q6600 dialed in comfortably at 3.6ghz as my workstation, it’s now time to build a new box to replace the aging home server, the Dell SC400 with a 2.0ghz Celeron Socket 478 inside. With 3gb of RAM, the CPU is practically pegged @ 100% utilization while running a multitude of services, in addition to 3 instances of VMs on VMWare. Hopefully, I can get the E2180 to the same 3.6ghz range without much trouble. Such a setup should have enough muscle to comfortably take over the same services, with maybe a few more VMs thrown in.

In any case, I am getting WAY ahead of myself here. I’ll report back next week with the results.

Microsoft’s ReadyBoost: ReadyGimmick?

Short answer?  Yes.  If you don’t believe me, then read this article.  Granted, the technology was developed while memory is still somewhat pricey.  Nowadays though, you’d be much better off running at least 1gb (most of us run 2gb) of RAM– at which point, there is absolutely no point in having that extra stick of memory funkiness sticking out of your computer.  All it accomplishes is suck up power.

Q6600 G0 overclock success

I put together my workstation replacement over the weekend (finally).  It ran prime OK overnight, and now it’s sitting at 3.6ghz stable.  Love how it is dead quiet as it sits 2 feet away from me underneath the desk.  Total cost for hardware?  Let’s break it down:

Antec Solo case: $70
Coolmax 500w PSU (the 120mm fan really keeps it quiet): $30AR
Samsung SATA DVD burner:  $30
Q6600 G0: $300 shipped
Coolermaster HyperTX2: $16AR
Abit IP35-E: $55AR
OCZ Gold 2gb: $55AR
Maxtor 500gb SATA: $107  after taxes f/ Fry’s
eVGA 7900GT: Bought for $180 2 years ago
Zalman VF900: $35
Logitech G15 Keyboard: $35 AR
MS Intellimouse Explorer 4: Can’t bear to get rid of this thing.  All the new mice are crappy anyway.  Bought new for $12.

Total: $925 on hardware, counting everything.

Now let’s count software:

Ubuntu Studio: Free
Salvaged copy of Win2k for Win Apps: Do I have to count this?  This was paid on some other system.  Let’s just call it free.
Google Pack: Free
Rest of the Apps: Free (Firefox, Audacity, Paint.Net, Visual Studio Express, VMWare player, VLC, IM Proggies, PuTTY, etc…)

Total Software Cost: Free.

Gotta love price wars.  Who would ever imagine 20 years ago how fast and cheap things would get??

iMacs in the enterprise

PC World published a piece by Rafael Ruffolo on corporations considering iMacs.  Here are some quotes:

According to Q2 results from IDC, in the overall U.S. computer market, which includes desktops and notebooks, Apple is now tied for third with competitor Gateway at 5.6 per cent of the total market share.

Tim Bajarin, president at Creative Strategies, said that while Apple lacks the enterprise sales force, servicing capabilities and the expressed interest in the enterprise sector, he does admit that this could change in the future.

My opinion is this:  Yes, this could change in the future, but not in the foreseeable future.   The reasoning is pretty simple.  First of all, Apple is not serious about competing for the enterprise.  Jobs is good at building machines that tug at the heart strings, but he is also VERY good at building devices that tug (pretty hard) at consumers’ purse strings.  Enterprise IT is about providing a pre-determined service level with a controlled budget.  For large enterprises with pre-existing investment in the MS architecture, migrating to the iMac does not make sense financially.

Neither does it make sense architecturally.  Apple currently lacks the will, or a way in providing a robust roadmap in providing a solution that can rival MS’s management features.  This alone would create havoc in a well-managed enterprise environment. Are you aware of any SMS or Forefront equivalent in Apple’s arsenal?

Lastly, such a move does not make sense in a resource management perspective.  Not only will you have to replace a whole slew of machines, you will also need investment in training your legions of support staff.  Sure, there will be some who are already well versed in OSX, but this is by no means the norm.

What I think the article is trying to convey is that the trend of Apple devices sneaking into the enterprise may increase– but that is easy to see.  All personal tech items will eventually sneak into the enterprise– this by no means will only apply to apple products though.  Personal blackberries/smartphones/PDAs will soon creep into the managed environment.  Sure, iPhone and iPods are popular, but their place within the workplace is currently dubious at best.

As the article said, virtualization may be the joker in this whole deck of cards.  Yes, the technology will enable iMacs to act like a PC, but at that stage, what’s the advantage of choosing the iMac + virtualization solution?  So users can also run OSX and Apple software?  Circular reasoning alert– OSX currently needs a separate management strategy to thrive in the enterprise.  Unless an enterprise has a financially justifiable reason to choose Apple (design, art, etc), I would be absolutely stunned to see any kind of inroad being made by Apple through such grassroot approaches.

Great sensational journalism though, PCWorld.  I thoroughly enjoyed the read.  This may just be the fuel that this small fire needs to create a more competitive enterprise IT marketscape.

Dual core vs. Quad core for VMs

Tom’s just posted an article on the benefit of going 4 cores vs. 2 cores.  Interesting read, however since the readership is heavily biased toward gaming, its testing apps were mostly games that have not been written to take advantage of the 2 additional cores.

As far as using quad core for virtualization however, the advantage is obvious.  Of course, performance is more dependent on the virtualization architecture than a few mhz in speed, as can be seen here.  If performance is a concern, then be prepared to throw more money on hardware, or stay with paravirtualized solutions like Xen.  My personal recommendation is OpenVZ / Virtuozzo for best performance as they are OS based solutions.  For manageability though, VMWare currently reigns king.  Stick with VMWare server if you want free, but keep in mind that performance will suffer.

Or I guess you can just throw more hardware at the problem like always.  Thanks, AMD & Intel for the price drops!  Please keep them coming!