The Lone Sysadmin

Mendel Rosenblum is one of the founders of VMware, chief scientist, and husband of VMware CEO Diane Greene. As such he’s got a lot of pull, getting his continuing vision of virtualization and systems management made reality.

He’ll be speaking in a few minutes out at VMworld 2007 in San Francisco. I’ll be liveblogging it in this post, so keep refreshing and I’ll keep posting.

8:30 AM - General session hall, Moscone Center halls B & C. Pop music playing, green and blue lights and graphics. As with the other liveblogging sessions a big thanks goes out to the VMware Community team for setting up tables, Ethernet connections, and power for us bloggers. Those guys are great and have really taken care of guys like myself. Thanks John, Robert, Brian, Anders, Cindy,

8:36 AM - A video of people at VMworld asking for virtualization in real life. “I’d like to snapshot conversations with my wife. That way when I say something stupid I can roll it back.” Hah! Awesome. Also talking about current wishes for VMware products.

8:39 AM - Karthis Rau, MC. Thanking folks for the party. Long tradition for VMworld to spend the last day on the future. Mendel Rosenblum on using mobility to transform your infrastructure, adding agility.

8:41 AM - Mendel Rosenblum. Going to talk about the state of virtualization, do some demos of stuff from VMware Engineering, and then we can predict for ourselves the future of virtualization.

Virtualization described as magic data center pixie dust. Not a helpful description. It’s more about complex machines interfacing with each other. You find an interface between them to interpose virtualization. As long as it still works like the old system you can get away with it. Adding levels of indirection, in CS terms. “All problems in CS can be solved by adding a level of indirection.”

(Funny, he’s nervous, got stuck, told a joke to break up the mental block. Awesome to see the human side of corporate leaders).

The first solution for the new layer of indirection was to cram more OSes on each physical piece of hardware. Second, because there is total control over where things are mapped we can do resource management and VMotion. Totally decouple them from the hardware to drive it as efficiently as possible.

Now we just adds pools of resources. The infrastructure makes the decision about how to use that pool most efficiently. Treat hardware as something you drop in.

8:50 AM - Rest of the industry really embracing this technology. An example: new hardware compatibility problems (making VMotion impossible to use). Causes partitions in the resource pool, makes life hard. Intel and AMD added capabilities to their CPUs to handle this, lie to the VM about the capabilities of the CPU so it remains backwards-compatible.

Talks about them asking for this compatibility early on. Asked the CPU guys to do this, they said “eliminate all the new features in the CPUs? What? No way!” And now we have it.

What sort of games can we play with this abstraction? Why can’t we move the storage around underneath the VM now? (my comment: SWEET, finally a public demo to confirm all those rumors out there). A demo of an ESX host with two data stores, unbalanced. Uses a script to connect to VirtualCenter and initiate storage VMotion. Graphs of the I/O to the data stores in real time showing the decline on one datastore and increase on the other. Relocation is complete. (applause!)

What does this get us? Leased storage is now easy to deal with. Capacity problems can be dealt with easily. Agnostic look at storage, you can move something from fibre channel SAN to a NAS device, no big deal. Excited about all these smart people working on placement problems for VMs. Now they can work on storage placement, too, optimizing that.

Next thing: virtual appliances. A review of the concept of virtual appliances, putting a stack of software together and building a “box” that you can download to do something specific, like CRM for your business. Gets rid of some of the traditional problems you see in an environment.

What about a marketplace for appliances? Appliances are really big, take a long time to download. What if you speed it up by adding “instant on” capabilities, so that you can stream a VM. Second demo, an engineer named Bruce showing the instant on appliances.

410 MB browser appliance, comparing the traditional download method to the streaming method. Starts a download on the traditional one, then goes to streaming one. “VMware OnDemand” the new product name, and a new stream manager built in. It’s prefetching what it needs, prioritize the blocks in the appliance that it’ll need. The appliance recorder analyzes the VM as you boot it and run some things, and then it orders the blocks appropriately.

Uses: marketplace, testing, ISVs, all sorts of things (my comment: ACE). Works with VDI, one of the big use cases.

9:03 AM - Software delivery models. Traditional (local only), SaaS (software as a service), hybrid (server-side management but local execution). See all of these models coexisting into the future.

Switches gears. High Availability. Describes a VM problem such that the infrastructure can notice the death of a VM and act on it, restart it. With the virtualization layer you can record the execution of the VM inexpensively, then take the log of execution and replay it to see where things went wrong. Present in Workstation now.

What if that is combined with HA to do Continuous Availability? Running a log of execution, sending it to a secondary VM to be replayed. (my comment: SWEET. Here’s my general DR solution. Plus, in a single stroke they killed a number of third-party products.). If there’s a problem with the primary VM or host the secondary can be activated seamlessly.

Demo from Dan from Engineering. An MS Exchange VM running LoadSim. Shows two VMs mirroring each other through the Continuous Availability feature. Mendel asks “If I pull the plug it’ll continue to run?” and goes looking around for the plug. Laughter. Plus the plug, the first VM dies, and in about five seconds the secondary takes over.

9:11 AM - Into the future. What are the hard IT problems to solve? Can take care of CPU & memory management now. New DR products now to manage site failures. New storage management tools. New hardware fault tolerance features. Can we guess what’s next?

Another thing to talk about is how the virtualization layer is responsible for the optimization of hardware. How do you do this in the most efficient way? In the future we can fine tune this process. We’re really conservative now, but as the algorithms get better we can do more. Power efficiency is a good area to tune for.

We’ve only scratched the surface with virtualization. Not much deployment, really, but as people see it the deployment tends to be rapid. Real bump up in efficiency. But it’s compatible with the way things are, so you can just slip it into your data center. Then you get opportunities to do new cool things.

9:16 AM - Thanks everybody. Karthik Rau again. Virtualization driving a complete refresh of the data center.

VMworld Europe 2008 is Feb 27-28 in Cannes, France. VMworld US 2008 is in Las Vegas. Powerpoint and audio for each session will be online.

Thanks all of us for coming. Safe travels.

(My comments: I won’t beat the pundits to the analysis but this is groundbreaking. More later).

Hey all. It’s been a busy couple of days here at VMworld, and over the next day or so I’ll be transposing my notes from the sessions into the blog.

If you missed yesterday’s keynote by John Chambers of Cisco (I did — my iPhone locked up and the alarm didn’t go off, which is under investigation) there’s coverage over at the VMTN blog, and links to a number of other guys who liveblogged it.

In a few minutes I’ll be liveblogging Mendel Rosenblum’s keynote. Stay tuned.

Welcome to my coverage of the VMworld 2007 keynote. We’re expecting interesting things today, announcements that will change the landscape of enterprise computing on both technical and business process fronts. Should be interesting. I’ll be updating this post, liveblogging style, as we go.

7:57 AM - Waiting to start. Loud rock music playing for the nearly 10,000 attendees. We’re in Moscone South, halls B & C. A big thanks goes out to John Troyer and the rest of the VMware Communities team who have set up tables with ethernet, power, and other friendly bloggers. :-) VMware rivals Apple in rabid fan base and it’s those guys who keep it going. If you’re out at VMworld 2007 stop by the VMware Communities booth in the Solutions Exchange.

8:07 AM - A theatrical performance on stage, with violinists. Lots of ETC Source 4s and Martin motion lights above the stage making them look cool. A video playing behind them showing the benefits of VMware and welcoming us.

8:09 AM - Karthik Rau. Four years since first VMworld. New things, multicore and 64 bit changing things. Network and storage boundaries blur. Virtualization moving from a tactical solution to the foundation of enterprise infrastructure. Thanks for coming to VMworld. The conference is for you, to share, to hear the thought leaders. 10000 attendees, 147 exhibitors. Introduces Diane Greene.

8:12 AM - Diane Green. Welcome. South hall widest aspect ratio room in the world. :-) Four years, even a year ago, this wasn’t an industry like it is today. Maybe due to IPO, thanks all the partners and customers.

Complete infrastructure refresh with virtualization. Better way to provision, DR, save power, etc. Saves lots of money. Allows people to save money and use that for rebuilding data center. Lots of new products.

Things are being optimized for virtualization. Processors, system management, applications even are getting tuned and optimized for virtualization. Great time to be in the industry.

8:15 AM - Stops to clear up the distinction between the hypervisor and the operating system. Also clears up that a hypervisor is not “virtual infrastructure” by itself. Infrastructure is everything else around a hypervisor.

8:17 AM - Explaining VMware’s hypervisor. Footprint of OS to support it (RHEL 3) is 2 GB. Hypervisor is 32 MB. Announcement of ESX Server 3i. Embeddable, OS-independent. “Absolutely no tie to an OS.” Easily installed, configured. Easy to deploy, easy to manage.

Working with hardware vendors. First to ship is Dell. Mark Jarvis from Dell to talk.

8:20 AM - Mark Jarvis. The demo is so easy that he thinks executives can do it. Challenges Greene to operate the Dell equipment on stage.

Dell system can take 2x the RAM, 4 I/O channels, new AMD Barcelona CPUs. In Dell 2×50 format. Designed for 40% better performance and 25% less power consumption. Diskless. ESX Server on flash in the hardware.

Demoing VirtualCenter. Greene adding a VM to VC (neat to see a CEO doing it), booting it on the server that they just fired up.

Standard interface that hardware partners can use to manage the hardware. “Health Status” in VC. Greene: “I didn’t think I had to think during this demo.” Laughter. Can manage the hardware now through VC.

8:25 AM - Video from Michael Dell (Dell), Bill Zeitler (IBM), James Mouton (HP), Barbara Schadler (Fujitsu), Yoshikazu Maruyama (NEC). They all said they’re ideal for VMware, think 3i rocks, drive value, yay.

8:30 AM - Video from beta customers. Faan Deswardt of Wyse. Kevin Hickman (Chevron), Peter Amstutz (missed it, sorry). Dramatic effect in security, virtualization operations, way easier for remote administration. Stop thinking of hypervisor as software, but more as a component of hardware.

8:32 AM - Greene. Working on standards, OVF, with XenSource and Microsoft, for virtual machines. Submitted it to standards bodies. Management of VMs is the challenge.

Site Recovery Manager: difficult to set DR up, terrible to test. Built a workflow around DR, integrates with vendors replication products. Example of this is Bowdoin & Loyola, doing DR for each other.

Virtual Desktop Manager: become a workspace to be accessed from anywhere, any form factor. Mobile workforces, distributed, seeing big advantages to virtualizing desktops.

8:35 AM - Virtual appliances. 600+ virtual appliances. Bea with a thin OS. McAfee moving infrastructure into appliances.

Energy savings. Everybody needs to save energy. EPA report mentions virtualization. PG&E gives rebates to customers for virtualization (and many other utilities doing the same).

Conclusion: hardware more virtualization enabled. Automatically handle faults, disasters. Simple provisioning, simple management. Automate your data center. Major steps to realizing the vision of virtualization. Thank you.

8:40 AM - Greene introduces Intel Pat Gelsinger, Senior VP of Intel. Recent investment, tremendous value.

9:41 AM - Pat Gelsinger. Alphabet soup of everything Intel is doing to help (I/OAT, etc.). Collaboration on standards, vSPEC, VMmark.

Graph on CIO issues in IT, business processes as much as technology. Environmental compliance, TCO of data center, security, productivity barriers. Deliver an IT strategy to support business challenges. IT must respond to business challenges. “Foundation is economics, QoS, agility, and biz alignment”

How can Intel make trust management, energy efficiency, communications, etc. easier for organizations? Common layer is virtualization. Simple, powerful to address Intel’s six “pillars” of challenges.

Virtualization disaggregates the OS. Breaks hard binding of OS to hardware. Creates opportunity for “data center” OS in the future.

Virtualization usage models. Static server consolidation - reduce CAPEX, increase utilization. Multi-OS workstation. Dynamic load balancing - rebalance resources with demand. High Availability & DR - checkpointing, BC, automated restarts of VMs. Mainframe and legacy environments migrating to new virtualization.

Intel FlexMigration, FlexPriority, Extended Page Tables in 2008, Virtual Processor IDs to avoid flushing page tables all the time. DMA Remapping, Interrupt remapping, Address Translation Caching (Directed I/O). Replumbing everything to support virtualization. Virtual Machine Device Queues, I/OAT. PCI-SIG Single Root I/O Virtualization.

Security threats - new threats emerge as new technology is developed.
Reliability - more single points of failure. Need to work to cover those.

8:55 AM - I/O Virtualization. Trying to achieve native performance again under virtualization. Intel VT for DIrected I/O. New logic layer in the north bridge, does DMA and Interrupt mapping. Shipping in 2007. “Was a weakness in the architecture.”

VMDq - networking layer in the NIC to classify and sort packets for the right VMM. Reduces CPU and increases throughput. 2x+ I/O performance for NICs.

Energy Efficiency. IDC chart saying data centers are expensive as heck. $0.50 is for power and cooling, for every $1 spent. Climate Savers initiative. Reduce compute power consumption by 50% by 2010. 100+ companies. VMware is joining Climate Savers.

Good power management being worked on for years, at the node level. Making those capabilities more available to VMMs. Also want to extend the view of power management to the data center level. Global optimizations of energy efficiency.

Reliability. RISC vs. CISC sales. He declared the death of mainframes in 1980-something. Why is big iron still around? Long tail of mainframes. How can we accelerate the decline of mainframes? (chuckles).

A single platform failure takes out a lot of VMs. Need reliability.

New 7300-series quad socket, quad core servers. Demos a Proliant. Graph of old vs. new, new is 30% faster. Yay. Lots of RAM, lots of I/O.

His stated single takeaway point: something very powerful happening.

9:07 AM - Diane Greene introduces Chairman and CEO of AMD, Hector de J. Ruiz.

9:08 AM - Ruiz. Once or twice in a generation something is conceived that changes everything in technology. Much more with less. Shifts control from IT vendors to IT customers.

Hardware vendors - either get on board with virtualization or get out of the way.

AMD Opteron the first 64 bit CPU. vExtend technology to help with VMotion. Green Grid movement. Responsibility and opportunity to help the environment. Must make good business sense, though.

Data center power usage in 2005: 45 billion kilowatt hours, $2.7 billion. Equal to whole state of Mississippi. Aggregate power consumption doubled worldwide between 2000 to 2005.

$0.50 of each IT dollar is for power and cooling. Going towards $0.70.

Own AMD data center - consolidated 117 servers into 7 ESX servers and 2 ESX “swing” servers. 79% less power, cooling.

9:17 AM - Ruiz. 5% of servers in 2005 for virtualization. 15% in 2010. 68% growth in spending, too. Virtualization, as Malcolm Gladwell would put it, has reached the tipping point.

We must not take the future for granted. Are we doing everything to make virtualization as good as it could be?

Lots of vendors shipping 4 socket AMD systems.

9:22 AM - New news: quad core Opterons. Asks Leendert van Doorn to the stage to talk.

Their 2nd generation of virtualization support, too. Nested Page Tables (Rapid Virtualization Indexing) speeding things up. Near native guest performance. Makes virtual machine monitor code smaller, eliminate shadow page tables.

Talking about reasons to virtualize, etc. etc. Virtualization rules. You can do DR, etc. etc. Basically third repeat of same presentation, different company, same choir/audience.

Uses a virtual machine for his grandmother in Amsterdam (not a bad idea, actually).

9:29 AM - Ruiz again. Virtualization is disruptive, etc., but still young. Thank you.

9:31 AM - Rau again. Congratulates AMD on Barcelona release. Talks about vendor stuff, sponsors, press marketing blurbs read by him. Most of the audience leaving. Thanks everybody.

Moment of silence for 9/11/2001.

9:34 AM - Housekeeping. ESX 3i on USB keys for attendees. Cool. Video contest on how VMware has rocked your world. Making efforts to reduce paper.

9:36 AM - Done.

Thanks for reading folks!

Hey all,

I’m sitting at the VMworld keynote right now, as it gets ready to kick off. Stay tuned and I, along with a whole bunch of guys, will be liveblogging it.

I’ve been in San Francisco for one whole day now, and I’m having a blast. Here are some suggestions if you’re in town and looking for something to do.

New Delhi Restaurant

I like Indian food. A lot. My friend Bob turned me on to Rogan Josh and I’ve never had it better than at this place. Really tender lamb, spiced very well, and the garlic nan was good. Some of the nan was a little charred, but that just adds to the flavor, in my opinion. They were really busy when I got there so I ate at the bar. They don’t have a huge beer selection, but you know, that’s okay. :-)

Sweeney Todd at the American Conservatory Theater

If you want to catch some theater while you’re in San Francisco check out the low priced tickets you can get from TIX in Union Square. I picked up a $24 ticket to “Sweeney Todd” at the ACT. Great show. There are some other shows I’d like to get out to see, too.

The San Francisco Giants

My friend Rich (also in town for VMworld) and I went over to AT&T Park today to see the Giants beat up on the Dodgers. Well, it was looking like a 2-1 loss for the Giants until Durham hit a 3 run homer. If you want to see a game there are 7:15 games Monday through Wednesday against Arizona, and the Muni (lines N and T) run right down there. If you’re in the Union Square area just head down to Powell Station and catch either of those trains.

We got 7 day Muni passports from the kiosk at the end of Powell St., behind the hot dog stand. Worth the $24 for getting around, that’s for sure.

Gordon Biersch

After the game Rich and I ended up at Gordon Biersch, which is a nice walk down the Embarcadero from AT&T Park. They are a brewpub and have killer garlic french fries. Let’s just put it this way: if garlic reduces the risk of heart attack I won’t have one until I’m 140, after eating those. Mmmm. Great brews, too. Their Marzen is great.

On tap tonight and tomorrow:

Pho Phu Quoc is on tap for dinner. I’ve eaten Vietnamese there for the last 12 trips and I’m not about to break my streak. Rich and I want to hit Anchor Steam if we can, sometime. There’s also word that Guster is playing the Warfield on Friday. That might be cool. In between, SFBC, Jack’s Cannery Bar, Thirsty Bear, and all other sorts of great stuff.

If you’re going to VMworld and staying in the Hilton, don’t rely on Google Maps for directions. It thinks that 333 O’Farrell St. is Redwood Alley. Instead, the hotel is actually very close to the Powell St. BART station, making transit to the hotel cheap from SFO ($5.15 rather than a $40 cab ride).

This is actually where you want to go (if you can’t see the map use this link instead):

View Larger Map

If you’re interested in taking the BART take the Dublin/Pleasanton train to Powell St. Takes about 30 minutes and is way cheaper than a cab. I made a sample schedule for myself at www.bart.gov.

On my list of things I didn’t need today: VMworld changed my hotel from the Handlery to the Hilton.

Normally this wouldn’t be a big deal, but I want to stay a couple extra days beyond what my employer is paying for. So I’d made reservations for the 8th and the 14th on my own. It was $194 a night, which I consider a bit expensive for one guy, but I wouldn’t have to change rooms then.

So I called the Hilton to see if I could get a room on the 8th and 14th. Sure, for $490 a night. Ouch.

Since I’d have to change hotels anyhow I opted to go with a cheaper option, the Air Travel Hotel, a couple blocks away. I don’t intend to spend too much time in the hotel anyhow, but for $380 a night I can hoof it a few blocks. :-)