Slashdot Mirror
VMWare Rolls Out Their Largest Product Release
opieum writes "VMware has launched Virtual Infrastructure 3.0 today which includes ESX 3.0 and a number of management utilities." Relatedly Jane Walker writes "SearchOpenSource has two authors that try to show why VMware ESX Server is miles ahead of Xen and Virtual Server. Discover what to watch out for when running ESX Server and how to avoid sprawl in your virtual data center."
... their new not-so-subtle advertising section.
A year ago i used the trial vmware workstation for a while, i liked it, but i wasn't willing to pay the cash to keep it. Just recently VMWare released VMWare Server which works on my XP Pro machine and appears to be a rebranded VMWare Server 5.0 that I used a year ago, for free.
It was thier GSX server product rebranded.
WORD
We've started to use more and more virtualisation systems at work -- the vmware solution is by far the most sophisticated and performant we've encountered - and the upgrade path to ESX server is always handy. Clusters are a virtual (a-ha!) doddle to work with once you pretty much virtualise everything (and the performance isn't bad either!).
Roll on more vmware products to make my life a happier one!
Global symbol "$deity" requires explicit package name at line 2. - If only $scripture started "use strict;"
Then was their GSX Server equal to their workstation product? And if so why are they charging $189 for a workstation product when Server now does the same thing?
By far, and I dont work for EMC or VMware, ESX server and virtual center are Bad Ass. There is nothing greater than 0 to minimal hardware downtime. Finally getting the moneys worth out of the hardware. Being able to place a box in "undoable mode" rocks! (think "oops that patch just hosed my sql cluster" "ok, i'm fine again"). Being able to deploy the same server via image and deploy one in 30 min. Adding disks on the fly and growing disks with 5 or less min of downtime. Facts: 1. ESX Servers are mammals. 2. ESX Servers fight ALL the time. 3. The purpose of the ESX server is to flip out and kill people. I once saw an ESX server flip out when a physical server dropped a flopy disk, and the ESX server killed the whole data center! (insert tounge in cheek) Not to mention the countless Beowulf clusters, countless.
For server relocation, with ESX server the answer would be to VMotion them off to another server. In real-time, they happily change their "physical" server, without missing even a single ping. (yes, I've done it, and do it all the time at work). ESX3 is supposed to have all sorts of real-time improvements on this process, allowing servers to auto-migrate themselves to less-taxed hardware, etc., etc.
I just replaced 2 old servers, 1 running Windows 2000 server, and one running Linux. I had an IBM X31 Pentium M 1.3x ghz notebook laying around, that had a lot of memory and a 7200rpm 2.5" drive it it. I installed a SATA PCMCIA card and am running my virtual machines off of an external SATA enclosure and drive.
Now I know what you all are saying, but the X31 works great, and is plenty beefy for the 2 servers it is replacing (a Pentium III 500mhz and an AMD 1ghz). The great thing about it is, it is absolutely quiet, it has its own 12" screen, keyboard and mouse (track-pad), and it has a built in UPS system. I have it hooked up the the same UPS that was running the other 2 servers, so if the power goes out, this thing will probably run a week without power.
The SATA external drive is fast, so that isn't an issue, and since it is external I place the drive away from the computer and sight for safety.
VMWare Server is great, and I really appreciate the price (free). I'm currently using Virtual PC for my workstation virtualization (testing, different environments during development, etc), but since I'm so happy with VMWare Server, I'll be switching over to VMWare workstation on my next upgrade. If a client ever needs serious virtualization I'll recommend they give ESX server a try. I think VMWare giving away their basic server is a smart move for them.
The really nice thing about converting my physical servers to virtual ones is how portable they are now. I literally can suspend my 2 servers, disconnect my external SATA drive, move it to a beefy machine, connect it, and resume the 2 servers on the faster machine; that's slick.
Ummm... If you virtualize a dozen physical servers, you've probably saved enough money for redundant VMWare host hardware.
I've been working in IT for just under ten years now and I hate every vendor out there. They all suck and none of their products work worth a crap. I'm sick of wasting my time chasing bugs and applying endless patches as new issues surface. However, VMWare is the one shining light in my shop. It does exactly what they say it does and it does it flawlessly. Every feature is as you would expect and (ESX) host servers stay up for months at a time. Never have we had to reboot a host to solve a stability issue. It just freakin' works. After you've fought so many other products for years, seeing VMWare software in action is enough to make you cry.
So, first the obvious stuff you know. It may have no value to you, but for doing live demos and development environments its sweet.
vmware workstation - for $$ you get an amazing desktop virtualization environment perfect for people who write drivers and core operating system software. Snapshots and things, complete control over memory, "frozen in state" debugging from outside the vm.
vmware server - free. On the desktop, it lets you run more than one pc at a time. Also can run on a server -- even headless. It can start with the operating system and automatically load the vm's at boot time. A conside side app lets you manage your headless server platform remotely.
Then you get into their Data Center environment.
Don't think 1 machine. Thinking 10 machines. You deploy your vm's across them, using your EMC storage arrays. You don't even have to know which hardware is running your vm. They can be moved around at will. Add a machine to the pack and you increase overall power. A machine goes down? So what? Migrate the vm. The VM's all run with the same "drivers" which are virtual.
Have you ever kept a server longer than you wanted because you didn't want to deal with reinstalling an entire operating system and all the software just to take advantage of the new hardware?
The problem with quotes on the internet, is that nobody bothers to check their veracity. -- Abraham Lincoln
Is VMWare shipping with VTx support yet, like Parallels.com? I've got WinXP, Linux and OpenBSD running under Parallels, and the performance seems far superior to VMWare.
As it happens, I can't always get by in my job using only free non-commercial software. Now, I have to assume that several other people here are in the same boat, and commercial software can provide value to them. Given those circumstances, I'd prefer seeing a debate about the relative merits of particular software packages, and discuss it, rather than dismissing a product because it costs money. And if slashdot happens to make a side profit, more power to them.
So here's 7 things I can say as to why I deploy all new systems as VM's:
1) Upgrading / retiring a server? Set up the new box, install VMWare, shut down VM on old server, copy files, bring VM up on new server - it never will know the difference (and this is without a SAN!) Got a SAN - VMotion the VM to a new server -0- (zero) downtime.
2) Custom app you only want to setup one and forget it! Great Plains, vendor platforms, your monitoring and cacti box. Set it up in a VM and let it live. You're never going to reinstall the box, so why put it on a box you may have to reinstall
3) Backups of a physical server suck. Think, with the box running, you can snap a fully functional complete disk image and move it offsite via nfs, cifs, ftp. If there is ever an issue, you roll back to that snap shot and it's just as if the server had a bad shutdown. No bare metal recovery that takes hours and hours. We're talking minutes (in a SAN enviroment).
4) Need a server to test something - create it! Setup anything you want in a VM - it doesn't care. Don't like it? Delete it! Need more power? Move it! Take it home with you for the weekend? Install player on your laptop and take the files with you!
5) Big hardware is better hardware. Running an enterprise on comsumer gear with a special sticker on the front is just bad. Enterprise grade servers are beaten into submission and have the best possible components. Dell has been known to hault production of a platform if a vendor's component fails during testing (the PE 4400's had this issue ~4 years ago). Using VMWare you can buy 2-3 big servers, rather than the 5-10 pc servers. Get 8-16gb of RAM per system. Get larger hard drives, and not waste so much space.
6) Isolate those apps. Sometimes its just better to let each application server have it's own OS instance. That way if you ever need to, you can replace them without having to worry that some interdependancy on the box will cause failures.
7) Its good to be green - think of the power savings when your entire enterprise is running on 1/10th the hardware. Using a performance SAN and a bunch of DL585's I can't think of a company under 10,000 people who can't run off of 1 racks worth of servers. Think about it - thousands of users, 100 server, in one rack. I have clients that are in the 50-100 user range running on 2 DL385's or PE2850's.
As one of the developers of both VMware Server 1.0 and Workstation 5.x, let me clear this up.
(Also, this blog entry might help with a few common misconceptions)
VMware Server, while similar in appearance and sharing much of the same functionality as Workstation, is a completely different product with a different use case and target audience. It is the successor to GSX, and is for people who want to set up, well, servers! The key feature that Server has that Workstation does not is remoting, where you run a server on a computer and connect to it from a separate computer via a remote console or web interface. The VMs can start up with the computer, shut down with it, and can be accessed by multiple users. The VMs also don't require an X installation to run the VMs, nor does it require any sort of UI to be running for the VM to run.
Workstation has a number of features that Server does not have. Among other differences, it supports multiple snapshots, teams of VMs (where multiple VMs can start up/shutdown together, can be in their own special network with custom NIC speeds and packet loss), and 3D acceleration in the guest (currently experimental, and requires DirectX in the guest for now). We have a lot in the works for the product, and the gap will widen.
The one difference that people seem to for some reason get upset over is the price. Workstation costs $189, while Server is free. People have asked me why they should get Workstation if Server is free. The answer is that you should get Workstation if it has the features you want. If Server is better suited to your requirements or budget, go ahead and get that. We're not trying to force you into buying Workstation, and we're in no way crippling the VMs. A VM made in Server should work in Workstation and Player just fine. Likewise, a VM made in Workstation should work in Server or Player.
Workstation is not somehow "better" than Server just because it costs more. It's a different product. Each has their own strengths and weaknesses. Yes, Server is free while Workstation is not, and part of this is because that's where mid-level server virtualization products were heading. Microsoft was considerably lowering the price on Virtual Server in an effort to hurt GSX sales. Xen, while not a huge contender in the enterprise yet, is free and good work is being done on it. Workstation, however, is unique enough in its dev/test features and still has value that we and our customers still feel is worth something. And you'll see that value continue to grow over time, just as you will with our other products.
I hope that helped you understand why we're still charging for Workstation while Server is free. Choose whichever product you like: Player, Server, Workstation, ACE, ESX.. They're all fine choices, and they all offer solutions to different problems. It's not just about virtualization itself anymore. It's about what you can build on top of it.
(Opinions expressed here are my own and are not necessarily representative of VMware, yada yada.)
Settle down there, Gordon...
Please stand clear of the doors, por favor mantenganse alejado de las puertas
Since our EULA didn't allow benchmarks to be published, most likely the only benchmarks you've seen are some that the Xen folks did for their SOSP paper, taken on a version of VMware Workstation from several years ago before the EULA was changed to forbid publishing benchmarks.
I understand that recently we changed the EULA back to allow benchmarking again. Let's see if the Xen folks redo their benchmarks or keep making hay by comparing with the old VMware Workstation 3.1 benchmarks...
[Standard disclaimer: I work for VMware, but I'm not speaking officially.]
Not too long ago I ran the discovery and benchmarking on a big project to move the large internet credit card processor I work for to either Xen or VMWare ESX server. From the first benchmark to the last stress test, Xen outperformed and outgunned ESX at every turn. Here's the kicker! We had paid VMWare engineers helping us to configure and tweak the ESX boxes. As for help from Xen? Well, I had the user's manual and a subscription to their mailing list.
Management
Sure, the VMWare servers had nice pretty management tools that were probably a couple hundred yards ahead of Xen's CLI tools, but this company doesn't exactly tolerate idiots. The unix guys here are more than capbable of migrating to Xen, compiled from source with a customized kernel, with no problem. The command line configuration and live migration utilities are more than adequate considering we already have SSH access to the boxes in the back. There was no need to change the firewall configs to allow us VMWare console access or anything.
Performance
I ran series of benchmarks for the following applications: MySQL, Apache, Lighttpd, perl and php. All of the bechmarks were ran on the same hardware, I just re-imaged the two machines multiple times. Xen won in every race. As a matter of fact, on the dual core Opeteron SunFire the Xen vm was a whopping 600 seconds ahead of the VMWare vm at running MySQL's sql-bench suite.
Stability
Xen 3.0 is more stable, IMHO than VMWare. Though neither platform crashed or hosed, the ESX box had a lot of trouble keeping time via ntp and had some problems with disk I/O.
Distrust
I reported the time problem several times to the VMWare techs assigned to our case, and they assured me that it was a host os issue. Funny that this article mentions that ESX < 3 has a problem keeping time with a 2.6 kernel isn't it?
Future
Later this week I'll be recieving the first Intel VT enabled server we purchased. I'll soon see if any OS or any kernel (including GRSec patched) kernels can be booted under Xen. If that is case, my company is likely is to purchase XenSource's commercial products.
Be Safe! Sleep with a Marine. Semper Fi!
Sure, you can probably plug the numbers into a suitably complicated equation, but it won't be linear and it won't be "obvious". The maxima won't be at the same place for different hypervisors, either. That's the point. If you use a single number benchmark, you can (almost) always find something product X does better than product Y. If you have the full behaviour of the system written out, vendors can't obscure things like that. It's good for the customer, as they can then see what product does the best with the specific characteristics they have in mind. It's also good for the vendor, because there's no pretense and no FUD (so the customers like you) and there's no denial (so the developers respect you).
Now, are ANY vendors going to do this? And I'm including Xen and VServers in this. Probably not. There are risks involved in being that transparent, plus costs. And even if the vendors all agreed it was a good idea, you think ANY of them would volunteer to go first?
This is not to diss VMWare. I respect them (as much as I respect any corporate entity) and this is just as true of the Open Source solutions. It's merely the practical reality that promoting a product through total education of the consumer is something neither party really wants. Customers want plausible denial if things don't work out, and vendors are not going to tell you to go to their competitors.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
But in this case, who cares? No one uses VMs because they need to run something faster. VMs have become popular because we have an excess of hardware performance and can therefor consolidate. Or maybe we use VMs as a test tool or whatever but never for performance senitive applications.
But still, it should be clear to anyone why a company would not want bechmarks of it's products published. They can explain their reasons forever but everyone will always suspect the real reaon is that the benchmarks don't look so good.
The cost of electricity 5 years down the line is probably going to play the biggest role in determining how much hardware you can run AND cool. Consolidating systems into VMs makes much more sense because a VM is not using its entire allocation of memory all the time, so it's easier to save on hardware and energy costs by having fewer systems - the VMWare mantra.
The only place I have seen that this is not true, is when you have a large compute farm, where each system is dedicated to running just that one job, which can take tons of memory (GBs) and always hits 100% CPU for thousands of jobs. In that case I would rather run without VMs because each machine is exactly the same as the next one and imaging a new system only takes a few minutes, localization and all.
"Beware of he who would deny you access to information, for in his heart, he dreams himself your master."