TechAgility

With over 60,000 CPU’s in this beast it certainly packs some punch? And if you have a task that needs some serious computing like this then you can actually submit a request to get your jobs processed - submit proposals here - how cool is that?

At 500 Teraflops though this is quite a significant increase in global computing power and it makes you wonder just how much further this can increase? What if someone was able to create and prove a quantum computer? how much would that influence these sort of projects?

I does make me wonder whether or not there might be some breakthrough with how we manage the input side of things? Here we are with this enormous Computing power and yet we are still effectively stuck with a keyboard?

The World’s Largest Supercomputing Cloud

I had no idea the Hubble telescope could see only 12 billion years into the past.

Frankly, I’d never really thought about telescopes looking into the past until Dr. Michael Norman, a researcher from UCSD gave me a basic education in astronomy - and explained the Hubble looks at celestial bodies whose light is just now reaching us. But it can "only" see 12 billion years into the past - and that was a veil he’d like to pierce. (I asked him what he did for a living, he said, "I simulate the universe." Trump that job description.)

……..

I was asked to give a keynote to celebrate Ranger’s opening, and this was only one example of the flood of basic research and science that will now be performed on the world’s largest open computing platform. Open? The facility was funded by the National Science Foundation, and is committed to providing large scale supercomputing as a service to any researcher or scientist within the US (submit proposals here). Ranger is built entirely on Sun - to dip into geekspeak for a moment, here are the stats:

• In around 6,000 square feet datacenter space, consuming less than 3 Megawatts…
• More than 4000 quad core Sun/Opteron blades, 120+ Tb of DRAM, running CentOS
• Delivering more than 500 teraflops computing capacity
• Jobs scheduled by Sun’s Grid Engine
• Interconnected by two, 100 terabit non-blocking Magnum switches (horns optional)
• Data managed by the Lustre file system, on Thumpers
• More than 2 petabytes of storage
• Managed by our hierarchical data management SAM-FS product, archived to Sun tape platforms
• With overall systems managed and monitored by xVM OpsCenter (the world’s largest installation).

Jonathan Schwartz’s Blog

written by dcaddick

I was helping build a Road Case for demos at MS 2008 Launches and other uses and while building a couple of VDI’s as well as BladePC’s for the Thin Client’s to connect to we found that roughly half of the devices were not responding at all to a simple RDP connection.

One point here is that although this was just a small Road Case for demo’s we had decided to try and emulate a proper enterprise environment as possible in that the systems had been setup as Multi-Homed with dual NIC’s so that there is a Private VLAN for Altiris imaging jobs as well as a Public VLAN for the connections and normal work, this enables BladePC’s, VDI’s and Thin Clients to be re-imaged without causing disruption to normal LAN activity.

So with that caveat in mind, what went wrong and how did we fix it?

Earlier we had discovered that RGS (HP’s Remote Graphics Software) was not connecting either and this was traced to the fact that ALL of the systems we had set up to run RGS had managed to bind to the wrong NIC.

Action: Disable NIC2, restart RGS Sender service - test, OK

Open Networks, click on Advanced, make sure the correct NIC is at the top of the order

re-enable NIC2, still test’s OK

reboot, tested again, still OK.

As a consequence we have checked back up with the Product Development Team and the feedback is that there is now an enhancement request in so that during installation of the Sender component it will check for NIC’s and with more than 2 it’ll ask you which NIC to bind to. There will also be a configuration to ensure the right NIC is bound.

So that was the RGS sorted, so what was happening to the RDP connections?

We could ping them, we could connect via Telnet on 3389, the Remote Connection box was checked, and the user was part of the Remote Desktop Users group…..  very strange.

Checked the Event Log and found some curious reference to TermDD and an error 50?

At one stage I thought it might be MS kb555382

But it finally transpired that it was this "The RDP Protocol Component "DATA ENCRYPTION" Detected an Error…" error message

Unbelievably the cause is: A potential race condition between the Icaapi.dll and Rdpwsx.dll dynamic-link libraries (DLLs) may cause the private certificate key on the Terminal Services server not to be synchronized.

It simply means the invalid certificate is deleted and it is recreated on the fly on the next reboot

Resolution:

To resolve this issue, follow these steps:

1. Locate and then click the following registry subkey:
   HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\TermService\Parameters
2. Under this registry subkey, delete the following values:
   Certificate
   X509 Certificate
   X509 Certificate ID
3. Quit Registry Editor, and then restart the server.  (Although this states Server - it can happen on XP SP2)

Interestingly enough there is also a mention of this very same KB article at VMware with reference to VMware Virtual Desktop Manager (VDM) 2.0 Release Notes

written by dcaddick

Well who would have thought it? is this Novell rising from the ashes? It will no doubt be an interesting year watching still further manoeuvrings in the Virtual space - although I wonder if the US does finally get in to a recession will this slow down the pace a bit?

Certainly VMware has a lot more competitors to contend with now than it did when the IPO was released last year and I see there share price is now back under USD$60 after the dizzy heights of $120 late last year….

Novell acquires PlateSpin

virtualization.info has just learned that PlateSpin, leader in the P2V migration market, has been acquired by Novell.

The canadian firm acquisition further boosts Novell visibility in the virtualization space.

Novell already has a major involvement in the market since the early days of Xen development in 2004, when the company was announcing the inclusion of the open source hypervisor in its SUSE Enterprise Linux.

After that first step another acceleration was provided by the interoperability agreement signed with Microsoft in 2006.

PlateSpin is a valuble acquisition target for Novell not just because of its flexible migration tool, PowerConvert, but also because of the other products in its offering: a capacity planning tool, PowerRecon, and most of all a new disaster recover solution called Forge.

These technologies will probably go integrated with the Novell management solution ZENworks, adjusted to handle virtual machines since end of 2006.

virtualization.info: Novell acquires PlateSpin

written by dcaddick