Recently in Dell Category

Dell released some details about their new 11th Generation PowerEdge line, and while of course it's more of the usual faster procs, more embedded hypervisors, and higher density and more power-efficiency, the thing that catches my eye is just how awesome they look.especially when you've got them all stacked together.

The new M710 Blades are looking pretty cool, too.  I'd love to have 8 of these arranged in an awesome renderfarm, grid, or VMware ESX Cluster.  By my count, there are 18 memory sockets, which means at least 72GB of ram for the max (assuming 4GB modules).

Still only two CPU sockets (can I get some 4-way action please?***), but with 6-cores (and likely 8-core procs on the horizon), that's still a good deal of compute-power in a 0.8U package.  Here's my favorite picture from Dell's 11G Flickr photoset (because I'm a geek about blades).

And believe me when I say that I wish I could wake up every day and look at a row of DataCenter equipment that looked this HOT!

*** In the immortal words of Vala Malderan, "Not everything I say is innuendo!"

I think Dell shipped us the wrong Blade Chassis.  When I look at support.dell.com for the M1000e, I see this picture:

But we got this.  I think we're going to have to return what we got to get the right product.  Boy it's going to be tough returning that 400 pound monstrosity.

UPDATE: There was some confusion about this post.  We got the right equipment, only Dell's picture on their support web site is wrong.  Some people just don't get sarcasm.

I cannot even relate the amount of excitement I've been feeling over the last 4 months, ever since I first heard about the new Dell M1000e chassis at a Dell Executive Briefing back in January.  We were finally able to procure one, and in less than two weeks (really, the same lead time as ordering a PowerEdge server), a big box of sublime Dell goodness arrived at the Huntsman Hall loading dock this afternoon.

First, and we knew this was coming, the Dell chassis shipped in one box.  Think about that.  We ordered sixteen blades, which makes for a fully populated chassis.  Sixteen servers with 16 cores and 32GB of RAM apiece.  Shipped.  In one box.  That just blows IBM out of the water (and really all of Dell's other server equipment, too, if you think about it).  Good job, Dell, on keeping the delivery "debris" to an absolute minimum.

We received the box 2pm, and by 4pm, it was racked and cabled for power.  All we have to do is figure out the networking configuration, and we're off to the races!

Well, obviously there was more to it.  A fully-populated chassis apparently weighs 400 pounds (look at the yellow warning sticker pic after the break).  We had to gingerly remove all of the blades, power supplies, and fans, which probably brought it closer to 70 or 80 pounds, which is perfect for your average IT team to lift and rack into place.  Thoughtfully, Dell included a few sheets of bubble wrap to lay the sensitive blade equipment upon while moving the chassis into place.

I was pretty familiar with the product, but there was one thing that caught me by surprise.  There's a pull-out-drop-down LCD panel with a little thumb-pad that allows you to browse the various components of the chassis and do some basic configuration and monitoring.  Total hotness.

We've got some reading to do, but by my estimation, we are 12 network connections away from being fully up and operational.  Well, that and actually installing OSes on the blades, but that's cake.

Many many more pictures of our Unboxing Ritual after the break.

This post definitely qualifies as a "super-techy-technology-specific" post, so if this doesn't interest you, may I interest you in something more interesting? 

So, we've recently received a ton of iSCSI storage for Project Bubblewrap, the Exchange 2007 implementation and deployment project at the School.  One of the main considerations when moving to iSCSI is that, for all but the most expensive iSCSI products, only 1-Gigabit line-speed is available.  This is in stark contrast to the readily available (albeit more expensive) 4Gps fiber-channel connectivity with more traditional SAN storage.

Cheaper is sort of the operative word when it comes to iSCSI.  Ethernet switches are significantly cheaper.  Ethernet interface cards (also known as Network cards) are cheap.  Ethernet cabling (e.g. copper) is cheaper than optical fiber.  All these things play a hand in iSCSI's tremendous popularity with smaller business.

So, 4Gps default line speed per port with Fiber Channel.  1Gps line speed for Ethernet, with 10Gbps available but astronomically more expensive.  Why can't you just use more than one gigabit networking port and somehow bundle that aggregate throughput?

Enter Link Aggregation, also known as 802.3ad Link Aggregation, which almost all modern managed switches support.

We want to duplicate the 4Gbps line speed of Fiber-Channel (FC) with Ethernet (IP).  You need four ports, four cables (CAT6 or CAT5e if you please), and a little bit of configuration.

NOTE: For all you networking experts out there, bear with me.  I'm learning as I'm going.

First, a couple of things to note.  Link Aggregate Groups, or LAGs, are treated as a separate, single logical link.  So all the settings (like spanning-tree protection from bridge loops and VLANs) are configured for each LAG, independent of the configuration each member port in the LAG.  For example, the LAG can be up and operational, even though the member ports are configured to be shutdown.  Crazy huh?

Because spanning-tree settings are per-LAG, if you are configuring a inter-switch uplink, it's important that you have the appropriate spanning-tree settings configured for your LAG before you bring it online.

So here's the nitty gritty.

From a console or telnet session to the Dell PowerConnect switch:

• console# configure
• console(config-if)# interface range port-channel all
• console(config-if)# shutdown
• console(config-if)# interface range ethernet g(1-4)
• console(config-if)# channel-group 1 mode auto (add those ports to LAG 1 (of 8), and use the LACP protocol to configure the LAG)
• console(config-if)# int range port-channel 1
• console(config-if)# no spanning-tree disable
• console(config-if)# flowcontrol on
• repeat the above on the partner switch and it's LAG ports
• console(config-if)# no shutdown

In plain-spoken English:

• Enter configuration mode
• Select all port-channel interfaces; remember, a LAG is treated as a logical port, hence the "interface" command
• Just to protect yourself from bringing down a switch or subnet, make sure the LAG is off before configuring.  Not that this happened to me before or anything...definitely not twice,either.
• Select the ports that you are going to use in the LAG
• Add those ports to LAG 1, of the 8 available, and use the LACP protocol to configure the LAG (auto)
• Choose the port-channel interface you just created
• Enable spanning-tree
• Enable Flow Control, which is a specific iSCSI recommendation
• This is already in English so no translation necessary
• The port-channel 1 interface is still the active interface, so just bring up the LAG.

There you have it.  4Gbps uplink between the two switches!  The legwork for this particular implementation sort of opens up the door for other networking bandwidth aggregation options, especially because as we get into really dense blade-based servers and Virtual Machine infrastructures.  The next step is to try and configure a LAG or NIC-team from the actual servers themselves up to the switches, but one thing at a time, people!

I like to visit Dell's website and "thumb" through their online catalog of desktops, laptops, and servers, because, well, it's FUN!  ****

But also because it helps me stay on-top of what the "state-of-the-art" is for those products, and maintain that operational vocabulary of various vendors' product roadmaps.

Yesterday, I was configuring a new Dell M600 series Blade, their 10G blades that I mentioned  in a previous post, and I came across a dizzying list of processor choices:

• Quad Core Intel® Xeon® E5405, 2x6MB Cache, 2.0GHz, 1333MHz FSB
• Quad Core Intel® Xeon® E5410, 2x6MB Cache, 2.33GHz, 1333MHz FSB
• Quad Core Intel® Xeon® E5420, 2x6MB Cache, 2.5GHz, 1333MHz FSB
• Quad Core Intel® Xeon® E5430, 2x6MB Cache, 2.66GHz, 1333MHz FSB
• Quad Core Intel® Xeon® E5440, 2x6MB Cache, 2.83GHz, 1333MHz FSB
• Quad Core Intel® Xeon® E5450, 2x6MB Cache, 3.0GHz, 1333MHz FSB
• Quad Core Intel® Xeon® X5460, 2x6MB Cache, 3.16GHz, 1333MHz FSB
• Quad Core Intel® Xeon® E5310, 2x4MB Cache, 1.6GHz, 1066MHz FSB
• Quad Core Intel® Xeon® L5335, 2x4MB Cache, 2.0GHz, 1333MHz FSB
• Quad Core Intel® Xeon® E5345, 2x4MB Cache, 2.33GHz, 1333MHz FSB
• Dual Core Intel® Xeon® 5130; 4MB Cache, 2.0GHz, 1333MHZ FSB
• Dual Core Intel® Xeon® 5148LV, 4MB Cache, 2.33GHz, 1333MHz FSB

Arguably, the CPU has the largest power footprint of any other component in a server.  When your power envelope is as emaciated as it has to be in an ultra-dense blade-chassis, then every last watt has to be accounted for.  And, generally speaking, heat-output is directly proportional to power-draw, you have to take into consideration the cooling of those hot components (and not new hotness, either).

Can you make the informed decision on which processor above has the smallest power-draw?  Does X mean "extreme", a qualifier that Intel is fond of for their desktop processor line?  Does L mean "low-voltage"?  But then what about "LV"?

How about from this list of AMD processors available in the M605 blade?

• Dual Core AMD Opteron™ 2214HE, 2.2GHz, 2X1MB Cache,1Ghz HyperTransport
• Dual Core AMD Opteron™ 2216, 2.4GHz, 2X1MB Cache, 1Ghz HyperTransport
• Dual Core AMD Opteron™ 2218, 2.6GHz, 2X1MB Cache, 1Ghz HyperTransport
• Dual Core AMD Opteron™ 2218HE, 2.6GHz, 2X1MB Cache,1Ghz HyperTransport
• Dual Core AMD Opteron™ 2222; 3.0GHz,2X1MB Cache,1Ghz HyperTransport

A shorter list (AMD's latest server-class processors haven't hit the mainstream yet), but certainly no less difficult to decipher.  Is "HE", high-efficiency?

I had to dig into Intel's and AMD's web\sites to find out for myself.  I was right that "X" does mean eXtremely high power-draw; 120W for the X5460.  And HE probably means "high-efficiency", since the 2218HE only consumes 65W. 

The winner in my book?  Intel's L5335, an awesome 2.33GHz with quad-cores, and only 50W!

Dell? If you're listening out there, put a Watt-rating next to each processor choice, gosh darnit!  I shouldn't have to go to the processor manufacturer's website to figure out the power draw per socket.

Intel? AMD?  Sell your Greenliness!  Just call it the Greeneon and get that over with.

**** I'm not ashamed to admit it.  Dell's 10G Blade chassis makes my heart flutter whenever I see it.  It is beautiful and angular!  It is the new hotness!

So, I'm actually down in Round Rock, TX, attending a Dell Executive Briefing, today and tomorrow.  I'm sitting in the hotel room catching up on my RSS feeds, and one thing caught my eye:

Dell continues the innovation in Blade technology (via BladeWatch)

First, I have to give mad props to the BladeWatch blog.  It's, generally speaking, an aggregation of a lot of the DataCenter consolidation news around the interwebs, focusing on Blade Technology, but also Virtualization of every flavor.  It has a unique slant, which I appreciate, towards "Green IT", something I'm completely in favor of and hold quite dear.

I'll be learning more about the M1000e and M600 line of Blades and enclosures from Dell tomorrow, but I'm immediately excited by what I see on their website.  Finally, I think Dell has a Blade product-line that can really compete with the IBM BladeCenters that we've been running for years.

Oh, and you can buy them now!  Girlz?  Why didn't we hear about this *before* we bought that latest round of R900s for Project Bubblewrap?!?  Oh, and they run more efficiently than the other guys.  SUH-WEET!

Check out Dell's Blade Server Product Page for the current spec.  Also, check out that sweet step-by-step (pdf) on the unboxing of the Blades chassis from Dell, HP, and IBM.  I can *fully* corroborate the almost bass-ackwards packaging and process for the IBM BladeCenter chassis.  Awkward!

Saw an interesting article up on the Windows IT Pro website about Microsoft and Dell partnering with (PRODUCT)^RED [via WikiPedia], which you may be more familiar with on the consumer products sides of things, with retailers like the Gap and Motorola licensing the (PRODUCT)^RED brand as a way of raising money for various African continent humanitarian initiatives.

Probably the most visible aspect of the brand is emphasis on HIV/AIDS policies and practices (one of Product Red's main principles).

Dell will be releasing a RED-branded hardware, and Microsoft will include a RED-branded version of Vista Ultimate, available only on that RED-branded Dell hardware.

As a former Student Health Educator at the University, I'm keenly sensitive to the issues surrounding the spread, prevention, and education of HIV/AIDS on College campuses. But, the pandemic of this horrific disease in the developing world (and in underserved populations in our own cities), means to me that there is so much left to do.

Hopefully, this announcement from Microsoft and Dell is just the vanguard of a legion of another partnerships with mega-corporations we all know and love.

Who knows? Is a (PRODUCT)^RED Wharton Lab, 100% comprised of these Dell computers in our near future?

Read more about the (PRODUCT)^RED announcement from Microsoft and Dell and maybe purchase a computer or two to support the cause [Windows IT Pro] [Dell].