Slashdot Mirror
How Well Does Windows Cluster?
cascadefx asks: "I work for a mid-sized mid-western university. One of our departments has started up a small Beowulf cluster research project that he hopes to grow over time. At the moment, the thing is incredibly weak... but it is running on old hardware and is basically used for dog and pony shows to get more funding and hopefully donations of higher-end systems. It runs Linux and works, it is just not anything to write home about. Here's the problem: my understanding is that an MS rep asked what it would take to get them to switch to a Microsoft cluster. Is this possible? Are there MS clusters that do what Beowulf clusters are capable of? I thought MS clusters were for load balancing, not computation... which is the hoped-for goal of this project. Can the Slashdot crowd offer some advice? If there are MS clusters, comparisons of the capabilities would be welcome." One has to only go as far as Microsoft's site to see its current attempt at clustering, but what is the real story. Have any of you had a chance to pit a Linux Beowulf cluster against one from Microsoft? How did they compare?
What do you call a cluster of Windows machines
when they Blue Screen?????
A Cluster Bomb!!!!
Service guarantees Citizenship! Questions Guarantee GITMO.... Amerika Uber Alles!
Licensing would seem to be the first thing that comes to mind.
Software costs for 100 linux machines are close to nill.
Software costs for 100 Windows machines probably won't be.
Granted I have read the licensing on the MS Clustering link but if it like anything else you'll need either a license of some kind on every machine.
--"Karma is justice without the satisfaction"
Anyhow, imagine how much you're paying in software licensing for a large cluster? For a univeristy project, this just doesn't seem to make sense.
One has to only go as far as Microsoft's site to see
Ah, so this is a typical Ask Slashdot then?
--saint
To start, it would take a few hundred dollars per box to put W2000 on it, since they presumably don't want you to just copy their evaluation version.
So unless they're willing to give you their OS for free, why would you even consider it? Suddenly your supercomputer cluster would cost like a real supercomputer... then you could have just bought a real supercomputer!
From what I understand from reading Win 2k Advanced Server's help section on Windows clustering, it is mostly for stability. Kind of like a massive mirror raid system. I really don't see any performance advantage if you're looking for supercomputer speeds, unless your measure performance by uptime. As a side note, what were you using for clustering? I'm currently doing a cluster using mosix for my school and it seems to be going nice. I'm just curious as to what gives the best speed performance on the linux end.
can't sleep slashdot will eat me
With their vaunted stability, and marketing
savvy, their new Cluster Product will be called:
The Cluster Bomb!
Service guarantees Citizenship! Questions Guarantee GITMO.... Amerika Uber Alles!
From Microsoft's site: "The Computational Clustering Technical Preview (CCTP) toolkit is used for creating and evaluating computational clusters built on the Windows® 2000 operating system."
Obviously, they are now attempting to compete with projects like Beowulf. It's probably all part of the M$ aggressive stance on Linux (and other competitors). The real question is, has anybody downloaded this kit and played with it. It's just a technology preview, so how mature is it in comparison to Beowulf or other clustering technologies?
GreyPoopon
--
Why is it I can write insightful comments but can't come up with a clever signature?
"from the please...no-more-beowulf-jokes dept."
too late
Sent from your iPad.
MCS (Microsoft Cluster Services) are designed for load balancing and fault tolerance, as where Beowulf Clusters (AFAIK) are more for distrubuted processing load for performance increases (massive threading). MCS works quite well, especially well on Fibre Channel and Brand Name Hardware such as Dells and Compaqs.
Simply put, it works well (but the cost is often an issue due to the cost of hardware in an enterprise) but it is not the same clustering you see with the Unices. E-mail me at my account if you have more specific questions.
My intent is not to start or participate in a flame war, but the term clustering simply implies different things on different OS'.
A beowulf cluster is not limited to Linux, it could run on top of any OS. I believe NASA did the original design work to be OS agnostic.
http://www.windowsclusters.org/projects.htm gives a list of current Windows clusters.
Finally, are you out of your tiny little mind? I wonder why M$ is so keen to help. There is no such thing as a free lunch, espically from M$.
Ok, you have a solution in place. It works. Some sales guy wants you to change your solution that works.
Make him convince you that the time and cost of the switch is going to gain you something.
Does your current setup not do what you need it to do?
Hello,
We run a MS cluster here. VERY big app... so big, I am loathe to name figures, because that would identify to MS just who is talking here...
But, we use MS clustering for our web app. Our setup is that we have a database server with 4 procs, and a growing array of web servers with 1 proc each, all of which use disk space on a SAN. W2K clustering manages the load balancing as well as allocating disk space out of the SAN to virtual partitions as needed. The original poster is correct; MS clustering is for load balancing, not computation. I have seen many times Microsoft sales reps don't have a clue of what they're trying to sell; they're just told from on high to replace Linux with Microsoft wherever they can. I think this is clearly a case of that.
My advice? Ask the sales rep to demonstrate how MS clustering will solve a common comp-sci problem with more MIPS than each box alone has. Point out that you're not running a web server or any such service on these boxes, but that they're for raw computation. Even better, see if he'll let you talk to a technician on how W2K clustering can meet your 'unique' (at least to MS) needs.
Now, for everyone else... Don't get me wrong. W2K clustering is a great technology for building highly performant, highly reliable, highly scalable applications quickly and easily. But it scales in the direction of millions of users, not millions of computations.
I am disrespectful to dirt! Can you see that I am serious?!
Clustering for windows requires Windows 2000 Advanced Server, and a great deal of patching. And with old hardware you are out of luck trying to run Windows 2000 Advanced Server.
Distributed computing for Windows has been around for a while though, Seti@home has been doing it for years.
Looks like Microsoft is busy being slashdotted.
Chances are the MS rep didn't understand MS clustering. He just knew that you had a Beowulf cluster and he wanted to sell you MS software so he figured he'd sell you a MS cluster, regardless of whether or not it would do what a Beowulf cluster could do.
However there is a server solution I saw demoed at a MS DPS I attended called Application Center. It allows you to manage your cluster and distributes workloads throughout the cluster.
Now, I'm not sure if you NEED this to take advantage of Windows 2000 clustering. The last time I worked with a MS cluster was under NT 4 and it was failover only. The load balancing was "faked" by a router that would just alternate which server the request was sent to.
(insert "yeah but MS is evil" comment here)
(insert "yeah but Linux Beowulf clusters cost less" comment here)
(insert "yeah but who wants to have to reboot your cluster all the time" comment here)
(insert "I wish the sigs were longer because that's a really good quote by Richard Feynman" comment here)
"For a successful technology, honesty must take precedence over public relations for nature cannot be fooled." -Feynman
My managers will only buy windows products as they have a site liscense with MS. They are looking into Linux a little bit because of the Terminal server w/ load balancing does not load balance and the clusterd computers do not talk to each other. The profiles on the 3 clusterd servers do not update each other at all. This was much better than the last attempt my boss did using an IBM pre configured configured box the whole cluster got a BSOD and corrupted a drive losing data for 3 days. People were not happy.
I can only hope MS's poor performance will make them switch.
Vote early. Vote often. Vote CowboyNeal.
an MS rep asked what it would take to get them to switch to a Microsoft cluster.
first the rep needs to prove that $199.00 per node for software fees has to provide major benifits over the Linux cluster. How many windows clusters can he list for you to call and ask about it? refrences, ones you can call and talk to the guys running/maintaining it. Show where microsoft provided increased profits or savings over an open alternative.
If they cant give you a dollar amount that shows increased profits or major savings then be sure to tell the rep that he shouldn't let the door hit him in the ass on the way out. It isnt MS versus Open anymore in today's economy.. it's what can get it done and save me money or can give me more profits... and this is what makes Open solutions win... microsoft can't give savings and the performance difference isnt enough to give profits that will more than overcome the added expense of Microsoft.
Get real numbers, talk to real people running real clusters on all platforms. if you have real numbers then you can make solid decisions.
Do not look at laser with remaining good eye.
Windows clustering works as advertised for the most part, but is expensive. Some exceptions include heavily loaded machine pulling from fiber channel arrays and NAS. Both of the network attached devices seem to have some problems. Driver issues? Don't know.
Haven't seen the reported "bsod round table" where one machine crashes, shortly followed by another and another. The problems we have seen is a single machine bsods, and the other machines in the cluster don't realize it's down.
If your already in the MS camp, it will work, it look at other solutions. I think they will be more cost effective.
"Science is about ego as much as it is about discovery and truth " - I said it, so sue me.
...with M$'s "Computational Clustering Technical Preview":
* PLAPACK package (open source software)
heh.
-JT
You can do a windows cluster thing, but it's still not as good even as Condor for Unix. All in all, I'd say to tell them to go screw themselves unless they want to give you money for a LOT more hardware as well as software, to make up for the fact that you're not going to be able to do as much with it. If MS wants to be taken seriously as a hardcore number-crunching OS, the bastards can EARN it instead of trying to bribe academics.
I've been looking at this a lot myself now, as I'm also building a cluster for use in a computational bio lab at Florida State. It certainly seems that Linux is the only way to go right now. In case anyone cares, my cluster right now is 16 nodes of:
Tyan S2460 with 2 Athlon MP1800+ processors per node
1 gig PC2100 RAM per node
20 gig 7200 RPM Maxtor HD
3Com Gigabit over copper Ethernet
low-end cheapass video and floppy, etc.
All in these really nice rack cases, with a big black 2001 monolith-esque rolling rack to shove it all around in. It cost just about $26,000 to build so far, but the plans are to expand it to as many as 512 nodes within the next year or so. Whee!
How good is any MS product in its v1.0 release?
Seems to me that historically, MS rushes a v1.0 product out to stem the tide of a competing product and then spend the next couple releases getting a "real" product out the door.
I have zero experience with unix clustering but would be suspicious of the MS offering until it has a chance to mature.
That said, with the three computers I have at my place (a p3 desktop, a celeron I use as a low grade server, and my p3 notebook) I'd love to be able to set up a cluster for encoding. Such operations will be the killer app for clustered systems IMHO.
www.lonseidman.com
Gives new meaning to the term cluster-fsck.
You are being MICROattacked, from various angles, in a SOFT manner.
I believe Indiana University has two hardware-similar clusters, one running Unix and one running some flavor of DOS. I don't have to URL but it shouldn't be hard to find.
At my last job, we had a COW (Cluster of Workstations) running all sorts of operating systems. Except Windows. Why? Because they won't run in a production environment for more than a few days without freezing or crashing, and the system administrators refused to babysit them. With Windows 2000, I've had my home machine run for upwards of 28 days without a reboot, but only if all the video drivers are stable and the machine is not doing too much at any given point (say, burning cds while watching movies and keeping my net connection above 200k/s). But so help you if a driver freezes. There's no way to reset them. Your hardware will play into your decision as much as the operating system, I believe, due to stable driver support.
In terms of performance, Windows kernels have pretty good latency compared to 2.2.x linux kernels, so running a full screen dos app might give very good performance, but there's a lot of overhead munching into your RAM, which is likely to be an expensive premium on older hardware.
Lastly, with Windows, I've never heard of doing channel bonding for ethernet (3 100TX cards ~= 1 gigabit), nor diskless booting that I know of. These can be really necessary for large clusters to keep maintenance down and performance up without buying higher end equipment.
Any connection between your reality and mine is purely coincidental.
but based on personal experience, Windows ME is pretty much a cluster.
'Same speed C but faster'
While I haven't been near a Microsoft Cluster in a while, I do remember a couple of things that really stand out about them:
The number of systems able to be part of the cluster is severely limited. At the time, it was limited to 2, but I'm pretty sure that has increased to a somewhat larger single digit number.
The number of applications available to run on the cluster is just as severely limited. Again at the time, there were exactly zero applications, but I know that there is at least one (Exchange) now.
Given the limitations of what uses you can put an MS cluster to, I wouldn't bother with it in the first place.
"Suppose you were an idiot..... And suppose you were a member of Congress... But I repeate myself."
That MIGHT make sense if he didn't need hardware too. Since THAT'S the case, he should head out to IBM, get a decent midrange, and cluster linux on virtual machines.
No worrying about Software licenses, AND "Professional" support from IBM for both hardware and software.
"I can't give you a brain, so I'll give you a diploma" - The Great Oz (blatently stolen sig)
there are some (as far as i understand) very good macintosh clusters that are very easy to use and very fast. especially if nothing (significant) has been done yet, a macintosh cluster computing G4-optimized code would blow away anything else in its price range. I can't say I have ever used one of these, or any other cluster for that matter, but the genuine power and versatility of the mac tells me its gotta be good.
We researched MS Clustering very extensively. We're already an MS shop and even still it was cost prohibitive.
Notes from experience:
1) Clustering with Windows requires one of the following OS setups: Win2K Server WITH MS Application Center, OR Win2k Advanced Server. (Similarly with the XP platform)
2) OS Licenses therefor will run between $1000-2000 _per-machine_!
3) If you need Application center, which you likely will, you're talking (If I remember correctly) about another $1g per.
4) Of course MS is just getting into this so don't expect it to be easy, well documented or stable.
Finishing Notes:
Obviously, Linux would be mucho cheaper
Easiest, and still cheaper than MS would be the Plug-n-Play Mac solution!
No Comment.
Persuade one of your mates to sell them a site license for Linux. If that doesn't work, find some pro-Linux company and offer them some easy cash.
if microsoft could do a computational cluster, wouldn't they have a 5000 box cluster in the super computer list just like the linux clusters.
Check out ACME at Perdue University. It was setup by a couple grad students on the cheap and really is a model of inexpensive high-performance computing. I think they only spent a coupe grand on the whole thing with help from the school scrap yard. Some good lessons in there. Oh, and they run FreeBSD which, as it's name suggestes is FREE!!
Have a Happy.
This troll has a point, most Universities have a licensing deal with Microsoft that makes licenses much, much cheaper. So the OS cost should be inconsequential, what matter more is how it performs doing the tasks you cluster computers for and how hard it is to write code for that platform vs. Beowulf. I would think the CPU overhead of Windows would make it not fare well when compared with Beowulf Linux. I realize Windows can arguably outperform Linux on web serving but that's much less an issue of maximizing CPU use.
The only reason I can think of to go with the Microsoft clustering is if its the only way you can get some decent hardware. If you have 20 PIIs now, I would think 20 P4s would trump them even running Windows.
I've been lucky enough to have access to a Beowulf in my current job, and the way we use it, Windows just wouldn't fly, because there isn't a powerful command line.
We're primarily using the Beowulf for computations which are "embarassingly parallel" - in other words, tasks for which it is trivial to partition the input into 16 equal-sized pieces, give one to each node, and then collect the results and paste them together. For example, multiplying incredibly huge matrices and brute-force keyspace searching are embarassingly parallel.
For us, the primary advantage of running Linux on the Beowulf is that most of the time we don't need to write custom software to speed up a calculation. We just write a shell script that rsh's into each box and runs a program with slightly different command-line parameters on each one.
Obviously for some computational problems it's worth using MPI to have the processes communicate with each other, or load-balancing software so that we can run lots of smaller, but different-sized jobs, and these techniques would probably work equally well whether you're running Windows or Linux.
But for experimentation and prototyping, and quickly distributing easy problems, I think there's an incredible advantage to having a command line. (Of course you could install Cygwin on all of the Windows boxen...but why?)
There are plenty of resources on the net that provide specific details about building clusters and how to optimize the performance. don't forget applications need to be re-written to make them friendly to distribute/parallel processing.
If there's one place where Linux excells and Microsoft needs to be kept out of with armed guards, constentino wire, and rabbid dogs it's the computing research centers in higher education. Scraping by to live and make post graduate tuition can suck, but having to fight for grant money that only lines the pockets of the richest man on the planet just so you can do your thesis is adding way too much insult to injury. For the sake of future scholars, show this salesweasle the door with the help of your foot.
*** Sigs are a stupid waste of bandwidth.
Everyone talks about setting up beowulf clusters. It's pretty easy to set them up, just make sure there is a lot of usable bandwidth between the systems.
The question here that isn't being asked is about the application. Sure, you have a cluster. But just what is it doing? What numbers are you crunching with that many gigaflops? To take the beowulf idea out of the realm of geek bragging rights into actual useful production takes an application, and you can bet that most are customer designed in house.
Very little of the OS itself is involved in the real applications that make beowulfs useful and money-making. Take a look at your intended application, and see what its requirements are. If you are writing it in house, tell the MS rep to take a leap, since you wont have to worry about 100+ MS licenses, Visual Studio licenses, or whatever else. If your intended application requires MS OS underneath, hold out on the rep until he agrees to a dramatically reduced price on the software. But worrying about the OS in a cluster before looking at the application is counter productive.
Toodles D. Clown
For a computational cluster, the OS itself shouldn't really matter. What matters is, do you have the tools you need, and does the environment allow you to work with the cluster in a flexible way.
For a typical compuatational cluster, what determines the performance will be the quality of your application. Only if you pick an OS with some extremely poor basic functionality (like, horribly slow networking), will the OS have an impact on performance.
People optimize how their application is parallelized (eg. how well it scales to more nodes). The OS doesn't matter in this regard. They optimize how well the simple computational routines perform (like, optimizing an equation solver for the current CPU architecture) - again, the OS doesn't matter.
So, in this light, you might as well run your cluster on Windows instead of Linux, or MacOS, or even DOS with a TCP/IP stack (if you don't need more thatn 640K
However, there's a lot more to cluster computing than just pressing "start". You need to look at how your software performs. You need to debug software on multiple nodes concurrently. You need to do all kinds of things that requires, that your environment and your tools will allow you to work on any node of the cluster, flexibly, as if that node was the box under your desk.
And this is why people don't run MS clusters. Windows does not have proper tools for software development (*real* software development, like Fortran and C - VBScript hasn't really made it's way into anything resembling high performance (and god forbid it never will)).
Furthermore, you cannot work with 10 windows boxes concurrently, like they were all sitting under your desk. Yes, I know terminal services exist, and they're nice if you're a system administrator, but they are *far* from being usable to run debuggers and tracing tools on a larger number of nodes, interactively and concurrently.
Last but not least, there are no proper debugging and tracing tools for windows. Yes, they have a debugger, and third party vendors have debuggers too. But anyone who's been thru the drill on Linux (using strace, wc -l
So sure - for a dog&pony show, windows will perform similar to any other networked OS with regards to computational clusters. But for real-world use ? No, you need tools to work.
Simple... ask for :
- Modifiable source code... essential for University level research.
- Blanket permission to publish research methods and results, including code.
- No NDAs that could limit a student's job oportunities. (i.e. "No Compete" clauses etc.)
- Free or low cost would be nice
:o)
Jonathan WeesnerIf you want to do some kind of renderfarming or number crunching across a network, why would you need *MANY* copies of win2k Server? I might have missed a point, but win2k datacenter is about load balancinglike bandwidth managing, IO requests, and uptime if one of the machines fails, etc...
:)) , but in your case I'd say keep with what you've got unless you get a buttload of funding and a very good reason to move to win2k (which I don't really see), because a datacenter plus admin will cost you in the 6 digits to maintain and license.
If the server holds the data and you have a potential of a lot of clients doing requests (thus I/O, Bandwidth, like a P2P crunching system to name a popular example) In that example, I don't see why you'd want to switch to microsoft if you got it to work on linux, you'll need to have a very good knowledge (or hire someone with) of Microsoft Server products if you want to move to anything more than a standalone server. Also last time I checked with M$ for that solution because I wanted a safer domain and maximum uptime, everything was doubled for 2 machines, I thought it would be a bit cheaper than that but heck, for the price of the Advanced server VS the standalone, with 25 users, you can get an extra tape drive and cheap RAID1 to mirror your critical drives (on a small buisness server)
So if you mention that you WISH you'll get donations, and you want raw computing power, instead of buying MS licenses, concentrate on the goal you try to acheive: distributed crunching power with scalable servers, so basically you'll need HARDWARE to crunch. (I still don't get why you'd NEED server to run number crunching, workstations can do the same and transmit to a server, like I was stating before). Check what you have, check what you need, design around that, do a cost analysis since it seems to be very critical in your case.
There are some cases where you'll want MS servers, here at work I've setted up a MS server to have less configuration and troubleshooting issues with my win2k Pro machines (at least I know when something screws up it's MS related for sure
--- Metamoderating abusive downgraders since my 300th post.
Much of the comment pool seen thus far is about the MSFT view of clustering, which is entirely possible what the MS rep was thinking. Computational science simply is not on MSFT's radar as far as I can tell. Thank your favorite deity for that.
Here are your issues. Most computational science codes are written for unix based systems. Cluster codes are either latency sensitive (so they use Myrinet/other) with protocol stack bypasses, bandwidth sensitive (need hundreds of MB/s to do their work, or dataflow sensitive (need packets arriving in the right order in a just in time mode). Most computing centers have hundreds or thousands of users, who need simultaneous access to the resources.
The pragmatic view is that your cluster needs to support many users simultaneously (windows cannot), manage large data flows correctly over high bandwidth low latency pipes (windows cannot), and do so in a manner whereby your costs (porting code, end user costs) do not rise as the number of users increase.
In short, unless MSFT donates lots and lots of new hardware (256 nodes or more of late model AMD/Pentium gear, with 2 GB ram, and 50->100 GB disk per unit, gigE adaptors, etc), you very likely could not effectively run windows in the first place, and your costs to run would skyrocket without some serious software license donation by MSFT. Not to mention the cost of the programmers you would need to port the Unix codes to the windows compilers. Not to mention the additional support headcount you would need to maintain the beast.
In short, MSFT based computing clusters are simply not viable. This is from a cost wise basis, a time wise basis, a headcount basis, and so forth. If MSFT is willing to help you offset all the costs you would incur, great, go for it. Otherwise, have a good long look at some of the cluster linux distros, and stay far away from MSFT products.
Disclosure: I work for a cluster vendor. We will sell what the customers want. Customers do not want MSFT clusters. If they did, our business would be brisk. Never had one inquiry.
pretty much all of the Intel server cards as well as several of the desktop cards support channel bonding. all compaq server NICs support this as well, and it works great.
however, I would take issue with your assertion that 3 100mbit cards are roughly equal to a gigabit card. while it's true that something like 4 100mbit cards will give you close to the real performance of a gigabit card when used on a low end PC, there is much to be gained by using actual gigabit (use of giant frames, better latencies, etc.)
if you're going to build a cluster, and you actually have a budget, you're going to buy decent yet cheap server boxes. these will most likely include 64bit PCI slots, and there lies your motivation for gigabit. the performance there is unparalleled when using a real wirespeed switch, without using faster technologies of a proprietary nature.
my 2 cents.
EOM
resistance is futile...beowulf jokes are unavoidable.
on a more serious note, are any of the distrubuted computing projects (SETI@home, AIDS@, Cancer@, the gene folding project, etc,) using windows?
no? maybe there's a reason.
FreeBSD for the impatient.
Spot on, except for one thing. You don't have to run X on Linux, you do have to run explorer.exe on Windows.
blah blah blah blah Linux cluster blah blah blah.
...Execute search MS - terms: cluster
Results: Microsoft Clustering, formally known as wolfpack.
...Execute talk: Yes...MS does clustering, what would it take to convince you to use ours.
I think if I was in the customer's position, I'd agree to it just to shove it back in their face when I ask how it distributes the computing load etc...of course that would be
blah blah blah computing load blah blah
...Execute search MS - terms: computing load
Fuzzy Logic Results: Microsoft Clustering, formally known as wolfpack. Use for load balancing.
But your only real choices for what you want are:s eed/appleseed .html
Linux cluster (Beowolf)
http://www.beowulf.org/
or
Mac cluster (MacOS or LinuxOS, AppleSeed):
http://exodus.physics.ucla.edu/apple
There is no real, available, and viable WindowsOS solution. You would be in beta-land, at best.
Consider it if MS will pony up big time (all the HW, SW, and maybe cash for overhead, salaries, gifts to the campus library). Consider it a research project (proof of concept).
You won't be getting your "real" work done with this, though. Use a proven solution if that's your goal.
that he says he "works for a mid-sized mid-western university" when his handle has a link to a Ball State University email address.
Just come out and say it.
"A plan fiendishly clever in its intricacies"- Homer Simpson
I guess the topicstater had his answer; Beowulf is for fast computing, Windows clusters are for load balancing.
What I would find more interesting, how is the stability managed?
High end clustering on Intel is still delivered by Calderas UnixWare (Openunix now). A cluster consists of 2 rootnodes which manage the cluster. If one rootnode goes down, another clustermachine becomes rootnode.
The rest of the system is also designed that there is the least possibility of a single point of failure.
I do not know if Linux and Windows can compare to UnixWare clustering.
On Linux it will not be Beowulf then, Beowulf is for speed, not stability or reliability.
On Windows I am not sure, but I do remember Bill Gates at some ComDex event standing by a cluster of 7 Windows 2000 machines, running Fords website. He turned off 1 machine the hard way, and the website kept on running. I do expect him to exactly know which machine to power off, and be sure not to hit a rootnode.
Well, don't worry about that. We can get you back before you leave. (Dr. Who)
:)
hawk
XP barely functions on a Tualatin Pentium III. I wouldn't bring it anywhere near my P2....
/Brian
Quasi-related. Lets say I have a bunch of machines sitting around. Is there anything the end user would use Beowulf for?
I don't mean - would it make vi run faster. I understand the very basics at least. But aside from having to write my own code to take advantage of the cluster, are there apps already beowulfized?
I could see where POVRay would be a good app to do this to. Have the controlling machine ship off each line of the rendering to a machines CPU and come back with the result a lot faster.
With enough machines that were fast enough, you could almost provide a real time rendered view of a work in progress.
IAABA. (I am a beowulf admin).
Beowulf clusters get built to support your application, not the other way around. Your choice of hardware and OS will depend on the parallel nature of your code. Do you need myrinet, or can you get away with fast ethernet? Will your code even compile under win32? Do the supporting libraries (PVM/MPI/BLAS whatever) run under win32? What about the queuing system?
How are you going to manage the cluster? You need automation, even for small clusters. How easy is it to add a new user, apply a patch or change a bios setting on your cluster without having to plug a keyboard and monitor into each node? What about central logging? How about automated OS installs when you add another 100 nodes when you get your funding?
Oh. Benchmark, benchmark, benchmark. That means your code, running your datasets, on your hardware and OS. Not vendor supplied numbers. If you have a serious hardware vendor, you should be able to wrangle demo mechines off them. Try before you buy.
All site licensees should remember who it is they are dealing with.
In the not too distant past:
At least one cancer research center had its budget seriously dinged when MS educational licenses ceased to apply.
Large corporations everywhere were pressured into Enterprise Agreements requiring them to keep current levels of Microsoft software.
And so on.
The Enterprise Agreements are especially interesting if you remember the original plans for Windows ME. Microsoft originally planned to strip out all network support in an effort to force businesses to upgrade to W2K.
They ended up backing down that time.
Now, however, imagine a plan that requires clients to upgrade their software but doesn't require Microsoft to guarantee that the upgrade will contain the same functionality.
Oh well. None of my business. People dumb enough to get into these situations deserve what they get.
Been there. Done that.
MS clustering is for load balancing and stability. As such, it does a reasonable job. Beowulf is for high-end scientific computing and does a resonable job.
To really do clustering well, you don't want either. You want AIX or Solaris, but you probably can't afford them. But, Linux clustering in load balancing style is developing quickly with IBM, TurboLinux, VMWare and Intel doing many interesting things. Beowulf is still cheaper.
In any case, though, for your situation, there's only one solution and that's Beowulf. Besides afford MS' licensing fees, you mentioned that you're running on older equipment. I sincerely doubt that those servers could run W2K Server in its standalone mode, much less in clustering.
Steven
Years ago, I worked at an ISP that ran partly on Solaris, mostly on Linux. A few MS reps came in to try and get us to switch to NT. We let them go through their routine, then walked them around the operations room, telling them the capabilities of what we had, and asking if NT would match them. The response was repetetively "no". When we pressed them on a few issues, they gave in rather easily. When we asked them why you couldn't bind another IP to an ethernet card under NT without a reboot, they admitted "lazy programming."
So, take the MS reps through the operation, tell them the capabilities. Ask them if they can meet or exceed them. If they say "Yes", you're either not using the real capabilities of your Linux machines, or they're lying.
steve
Oh, you're not stuck, you're just unable to let go of the onion rings.
It's a .gov so they have to use some contract we already have with IBM hardware and the MS site liscense. The director of my department is a big MS fan (even after he upgraded to XP on his laptop and corrupted the drive). I hope to be moving to a different department tho where I can possibly run linux on my desktop pc.
Vote early. Vote often. Vote CowboyNeal.
Actually, thats kind of the point, in XP you can't turn off the GUI, whereas on That BSD/linux box you can, thus using less resources for the GUID and more for what you will actually be using the box for.
So your post just helped prove the point. =)
Do you Gentoo!?
Do you run X on the 486... probably not.. if you do a full install of redhat 7.2 and run x, your looking at 2.8 gig install, and running x, the system resources are going to be real similar...
This is true a full install of Redhat 7.1 is big, but who says you have to do a full install. A minimal install, including X Windows can be done in less than 500 MB, with little effort. If you go with Debian, I think you can get it under 300 MB. I also beleive there are mini distributions of Linux that do X Windows in 50 to 100 MB. It is really just a matter of what you need as opposed to what you want.
"Our products just aren't engineered for security,"
-Brian Valentine,VP in charge of MS Windows Development
It all depends on what application it is you want to dedicate the cluster to, and whether tools are already available.
If you wanted a cluster to do animation rendering and were based on 3D Studio, then you'd probably want your farm to run Windows. Web work could go either way, with Linux being a less expensive solution. A custom coded app would probably be best implemented under *ix, though I could see situations where there might be some resources more readily available under Windows.
You should check out the AC3 project at Cornell University's Theory Center, which is "home to the largest Windows-based high-performance cluster complex in the world".
There are numerous machines, such as the 256 CPU Veclocity 1, that run MPI-Pro over MyraNet(?), that was one of the 500 fastest computers.
Windows is a very viable and high performance solution for running scientific parallel application, and you should order the $8.00 evaluation kit from MS and check it out for yourself.
I've developed for some of these systems, and have been very impressed. I've worked with Linux clusters too, but only on older, weaker machines, so it would not be fair to compare the two.
(Btw. all opinions here are my own, and in no way should be construed as those of Cornell or the TC).
One thing you might want to consider is administration time, scientists, who are already annoyed that programming destracts them from their real work, might not want to devote the time and effort to learn to and administrate all those linux boxes.
Anyway, if the MS rep is very eager, he might offer you some great deals. MS is very eager to be taken seriously as an HPC option.
Yeah I was helping a grad student the other day setting up a Windows cluster.. its not fun.. not terribly hard, but some of the requirments, and that stupid ass HCL that they have make things a bitch. I would think it would be more likely used for load balancing than anything else, sure you could use it for computational things, but the overhead that windows needs for the OS means that you would get less CPU power than if the system was a Beowulf one
"After all, we're all alike."
:)
hawk
-
Failover clustering. This is an OS service that servers like SQL Server and Exchange plug into that allows Active/Passive or Active/Active clustering over a shared SCSI/Fibre bus. In theory you could write your app to use this service but I think it would be overkill.
-
Network Load Balancing. This is just a software version of the standard kinds of NLB found in cisco boxes.
-
Component Load Balancing. This is the most suitable. It's provided by Application Center and it allows you to deploy COM+ objects on a cluster of machines and have the calls distributed according to the load on those machines. You can control the threading and lifetime of the objects and view the status of the machines pretty easily using the Application Center MMC plugin (or SNMP, I believe). You'd have to wrap the computational part of your application into one or more COM objects. Once you've done that then you can create and call those objects in the cluster as if it were one machine - the clustering is transparent to the client application. I played around with AC a bit when it was in beta for a project that I was working on. We didn't go with it in the end because the design of our application ended up not requiring it (we just went with hardware load balancing), but it seemed like pretty cool technology - if you're into the whole COM thing. It has a really cool rolling deployment feature where you can redeploy your components (and/or IIS application if you have one) to your cluster incrementally while it's still running.
Here's some links to docs on MS's site:Introducing Windows 2000 Clustering Technologies
Application Center home page
Component Load Balancing
Back when Beowulf started there were a few universities that had dual boot systems. I think they ran Linux half day, Windows the other. Windows NT and above have had implementations of MPI and PVM that have worked quite well for a while. MPICH and PVM are both freely available, the problem is that these systems do not integrate well with Visual Studio.You end up needing cygwin/gcc, which is basically Unix on Windows
Most clustering software today is either open source or based on open source and made for the Unix environment. So besides the obvious license and stability questions others have brought up Windows has limited tools and libraries.
ALSO:
There are problems with remote administration, being tied to graphical interfaces for such simple nodes. Wasted efficiency by running the GUI.
Finally Microsoft has a VERY limited knowledge base for this application of Windows nor do many HPC people know anything about Windows.
As you can see from the above, Linux on HPC is basically able to take those same horrible excuses for running Microsoft on the desktop and shove it down their throats.
Yes, Windows can do HPC, but why would you want to?
A few points:
You can set up any application or service to cluster & fail over if required, as long as:
Active/Active mode is more complicated, meaning instances of an application running on different nodes, all accessing the same data on disk. Only certain applications can do this successfully, e.g. Oracle, which does so by using a custom file system and effectively bypassing the Windows Cluster Service. Windows & most apps will normally throw a fit if there are clashing file requests from multiple nodes, since Windows caches file tables in memory and can thus lose track of the real situation on disk (bad news). I've seen it BSOD in such cases.
(this is not a
At my place we used to have some WLBS (Windows Load Balancing Services) systems.
this article without the obligatory Beowulf comments?
Escher was the first MC and Giger invented the HR department.
This wouldn't have happened if they had thier servers clustered to handle the load-balancing issues.
about that mac solution....
/ hp c.shtml
Yellow dog linux sells a cute little piece of hardware designed for clustering around PPC. very cute...maybe the best balance of cost effective and easy in terms of clustering that ive seen.
http://www.terrasoftsolutions.com/products/briQ
-jef
my understanding is that an MS rep asked what it would take to get them to switch to a Microsoft cluster.
You've got a golden opportunity here! Microsoft does it your way or they don't get the sale.
Let them know the nature of a cluster in a research project. Nodes will be swapped in and out. New ones will be added. Different OSs will be used. So tell them you want a copy of Windows for each potential node, licensed to the University and not to any individual node. Tell them you need full rights to install, reinstall, and uninstall any particular copy on any particular node. Tell them you will not accept any terms restricting the cluster to Windows only.
If you really want to play hardball, tell them you don't even want licenses, but bonafide user-owned copies of Windows subject only to the provisions of copyright. In other words, you don't want to be subject to any EULA. Then you'll discover how much Microsoft wants your cluster to be a Windows cluster.
A Government Is a Body of People, Usually Notably Ungoverned
"... toolkit is used for creating and evaluating computational clusters" I really hate to say it but M$ would call any cluster of computers a computational cluster.
What about sourcing issues as well. As I understand it, when building a cluster, the software that you run has to to either be custom or custom modified in order to take advantage of the cluster. With linux/OSS this seems like it wouldn't be too much of a problem, as (providing the requisite skill is available) you could 'simply' modify existing aplications.
However, with Windows and windows software you often do not have that option. Is the management of processes entirely handled at the OS level? It seems like that might be somewhat inefficent, as opposed to having the program handle at least part of the management. If not, are there ANY aplications that are designed for a Microsoft clustering environment?
Yeah!! What a stupid idea! Nobody would buy that!!
http://www-1.ibm.com/servers/eserver/clusters/hard ware/1300.html
So which one of us is going to tell the multi-billion dollar company that?
"I can't give you a brain, so I'll give you a diploma" - The Great Oz (blatently stolen sig)
Try looking at Pooch from Dean Dauger. http://www.daugerresearch.com/pooch/whatis.html
This would allow you to use the Macs (OSX UNIXY goodness too) individually as personal workstations (for writing, graphics, computation, surfing the web) while at the same time using them in clusters for compute intensive work. This makes for a doubly productive machine and one that is much cheaper as more work can be accomplished with it than simply using it as a dedicated node.
Mac clusters are easy peasy to set up (even junior high students are doing it) as the one page instructions should indicate and Applescript'ability. Also pretty damn fast given the built in Gigabit of G4's and the Altivec (if taken advantage of like in Apple's version of BLAST).
Finally, the other item of interest. You can use any Mac you have. G3's, G4's of any model and speed as one does not have to balance everything like on typical clusters where all of your hardware has to be exactly alike. The Macs in your cluster can even have iMacs on the secretaries desk involved!
Visit Jonesblog and say hello.
Then, point out the scads of Beowulf clusters and Linux/Unix based systems.
Finally, inform the rep and your management that you've chosen to use the more cost effective, higher performance and standardized choice...Unix.
If management resists further, do a cost analysis. That'll convince them.
299,792,458 m/s...not just a good idea, its the law!
Galileo: "The Earth revolves around the Sun!"
Score: -1 100% Flamebait
...to ask a question that I wanted to ask as well. Granted, this topic seems a little strange, considering the Linux cluster is in place, and it seems like the kind of question which encourages a Microsoft vs. Linux world domination showdown for grandmaster of the universe. It also shows a limited business sense on the part of the poster (why change something that works well when you can't afford a replacement?).
Right now a coworker and I are looking at pricing and configuring a fault-tolerant cluster for a client who runs Windows 2000 and Exchange 2000. They're a bit paranoid, so they've decided they want a cluster. We've tried to educate them on exactly what a Microsoft cluster can and can't do, so it's difficult to understand exactly what they want (basically an entire network exactly like Microsoft's own, but for $1000).
Pricing on a two system cluster is around $50,000. Buying two copies of Exchange and Windows Advanced Server will total $20,000. Then there's the hardware costs. For our client, they've specifically requested this, so they're ready to pay.
My question to Whamo is are they really taking the Microsoft rep seriously? If they have to pay software costs for their new cluster that's going to mean two things: either buying less CPUs to add to the cluster, or not doing the project at all, because just the software will put them over budget. With Advanced Server running somewhere around $4000 that's a lot per machine when Linux costs at most $5 to burn a CD after downloading it via the university's T1/T3/etc. Whamo says "it is running on old hardware and is basically used for dog and pony shows to get more funding and hopefully donations of higher-end systems" and to me that is your answer. If you can't afford the hardware you can't afford to buy Microsoft's software...
Also, there's MOSIX as well, but I don't have much experience with MOSIX and thus cannot comment on it.
Comment removed based on user account deletion
From the link to MS, it appears that although the versions of Win2k are evaluation versions, the version of Visual C++ isn't (I don't give a stuff about the rest).
If you can cope without the optimising compiler of the professional edition (which appears to make little difference which optimisation method you choose), $7.95 (+$1.95 to get it to the UK) for Visual C++ seems like a bargain.
Steve.
I've clustered Windows boxes plenty of times. I leave them clustered in a corner of a dark closet. Unplugged.
A cluster of Windows would be a Glass House, and you know what they say about those.
Quad Pentium 500 MHz Pentium Xeon's
1 GB memory
6 disk array, three logical mirrored drives.
MS Windows NT Server 4.0 Enterprise Edition
MS SQL Server 7.0 Enterprise Edition.
It should be noted that you must have two NT Server licenses and two SQL Server licenses. If you want to do an active-active cluster it requires four licenses. The Enterprise Editions of these software packages was much more expensive than their standard counterparts. You can not use standard editions for clustering.
Installing cluster services was very easy. The Cluster Manager app was OK outside of the occasional hangs. (Although the manager app hung, the operations were completed, such as failover, failback.)
In order to do active-active clustering you must have two shared storage devices; the active node will only be able to access the shared storage it "owns".
SQL Server installed all right if you followed the MS White Paper exactly. I don't know why, but installation order was important; if you didn't follow it it didn't work.
Applying service packs was extremely painful. The instructions were straightforward but did not work. MS provided us with a program that backed out the SP snafu, which worked somewhat. If it weren't for google we'd have been dead.
MS support is useless IMO. No contracts just pay-by-incident. Have a credit card handy before you do any upgrades of any kind. You will most likely need it.
As long as the cluster was just doing SQL server, it worked great. Failover was seamless. Given the proper hardware, Windows behaved well. Make sure that you only attempt this with certified hardware. Very important.
Once we started adding third party reporting software things started to go bad. Adding it to the cluster services was remarkably easy. However, even though the servers had quad procs and a good amount of memory, simultaneous report requests ground the system to a halt. SQL Server behaved well, around 25% of CPU at most even in heavy load. The reports (JRE) would take up over 50% of the CPU in light load. Very bogus IMO.
A lot of third party apps do not support MS Clustering. Lot's of tweaking to get them to work.
If I were to do it again, I think I would not have used MSCS, but instead have two distinct systems that had some kind of data replication software.
This configuration is also limited to a two node cluster. Although you can run an active-active cluster the instances of SQL Server would be seperate. The data storage areas cannot be shared between the two nodes.
Although I prefer UNIX I try not to be an MS bigot. It does certain things well. I hope that clustering has improved with w2k.
The interesting thing about MPI (which, by the way, is the clustering software that's part of Microsoft's clusting kit) is that it's also available on Linux, Solaris, and other OS's if I recall correctly.
We used it in college, and we clustered between an array of Solaris SPARC machines and an array of PCs running Win2k. If you wrote your software correctly, it didn't matter what OS was on the thing as long as you had a version compiled for that environment.
So what I want to know is why on earth anyone would want to build a Windows cluster when they could avoid the bloat and use the exact same software in a more robust and secure environment of Linux.
In the Windows environment our accounts had to be granted administrative access in order for us to run our projects, we'd request the access and it'd be granted for the duration that we were in the lab, then revoked when we left, even if we were just "going to the bathroom quick."
Slay a dragon... over lunch!
that rant would only be good for idiots who dont know what the hell a beowulf is for.
when was the last time you used VB for engineering, computationally heavy tasks?
christ - almost all this shit is written in Fortran still... Fortran 77, i believe and not even 90.
try to change the value of 5 in VB, go ahead... i dare ya.
... hi bingo
Then run the new distributed computing, called The Distributed Folding Project (http://www.distributedfolding.org). It just started last month and there are clients for Linux, Mac, *nix and Windows (in other words: most platforms in use today). I believe this is the first distributed computing project that has a Beowulf cluster install guide :)
If you have questions, visit the forum here: http://groups.yahoo.com/group/distributedfolding/
Ignoring your apologetic about the web page being old...
Interesting, so they got a 144,000% increase in speed? Was this distributed over 1500 computers? You don't get that kind of speed up going from one computer to 10 or 20.
Sorry but that sounds too good to be true. You aren't astroturfing are you?
As for the Web Services aspect... You aren't seriously arguing that PRESENTATION allowed them to significantly reduce their development time, are you? They had to cluster their algorithm to get a 1440x speed improvement and it was all worth it because web services made it easy to view? Huh?
Education is a better safeguard of liberty than a standing army.
Edward Everett (1794 - 1865)
Microsoft DOES have innovation, quality, and bug-free code, just not TRADITIONAL innovation, quality, and bug-free code.
50 Meg? Shoot, I have Jailbait Linux running on my I-Opener. On a 16-meg SANdisk. X, Netscape, VNC, the works.
As near as I can tell, you'd be lucky to fit the embedded version of Windows into that much space. This is a complete desktop distro, with all the yummy command-line goodness thrown in.
- fader
The combination of "Microsoft" and "clustering" is where we get the word "clusterfuck"....
Max
My god carries a hammer. Your god died nailed to a tree. Any questions?
A company out there provides commercial MPI libraries for a variety of operating systems, Windows included
http://www.mpi-softtech.com/
A couple years back, I was one of the MPI writers/maintainers for a number of platforms (worked with MPICH). As far as performance went with real applications (as well as synthetic benchmarks), Windows and Linux on the same hardware were pretty much the same. Typically computational problems were faster on Windows though because of the better compiler support at the time. Communications performance benchmarks were interesting.
Also, the at the URL above you can find cluster management software (batch scheduling and stuff).
Ugh. I am putting together a win2k cluster at my job, and I have their computational clustering technological preview. For the most part it's a MS marketing scam (Here build a cluster on these trial versions of win2k, and check out our awesome Visual C++. Oh and here're some old versions of the stuff you really need to build a cluster.) It's not really that great IMHO. All you really need is MPI and a bunch of windows boxes. MS likes to push the proprietary MPI Pro from MPI Software Technology.
The AC3 folks at cornell have done quite a bit with these windows clusters. I guess the parallel Matlab is pretty nifty, but there's no reason any of this stuff couldn't be done on a more mature platform.
Personally, my biggest turnoff is the fact that you need KVM switches wired up to each node...well that and the overhead of running the bloatware that is win2k. Compared to a 256 node headless linux cluster we built this just sucks. Hard.
I don't run explorer.exe on my windows box at all. I use an open source 32-bit shell called Litestep. It is infinitely configurable and themeable. There are tons of themes to download, or you can dive right in and edit the rc files yourself. You can make it look like any Linux WM or desktop environment. I love desktop-click popup menus, which is one of the countless modules available. The main litestep.net site is down right now, but checkout Shellfront for info and links on Litestep and a few other replacement shells for windows. If you know Win32 programming, grab the source and dive in, the dev team is in a bit of disarray at the moment.
chavo
It's nice to see how a little competition encourages Microsoft to innovate. Would they have even considered offering computing clusters without the competition from Beowulf Linuces?
The bitter lessons of a veteran coder: http://bitterprogrammer.blogspot.com
So, why in heaven's name would anyone want to run Windows for clustering? Oh, I can think of one reason: Microsoft may give you a big bundle of money, like it did to these people, but other than "buddy, the first one's on me", they just don't have a point, or a product. The only area where Microsoft is somewhat credible is on the desktop; their product isn't great, but it's usable. For server applications, they are simply not competitive.
I mean, it's not really an advantage to pay for licenses, throw away a lot of work you've done on the cluster you have, train for a new environment, etc. Other questions are: which applications are supposed to run on those servers, who is supposed to work with them in the production phase, will you develop for the thing and with which tools?
For the MS-person it'd be great if you switch, so he can tick off another customer converted to Windows, but what do you get out of the deal?
--
"By the way if anyone here is in advertising or marketing... kill yourself." -- Bill Hicks
VMS clustering *IS* the best implementation. But Windows clustering is nothing like the VMS version. Microsoft got Dave Cutler to reimplement the core VMS internals, but they failed to hire the cluster and file system people from DEC.
My experience is with Windows NT 4 Server Enterprise Edition. MS chose to use a "shared nothing" implementation - which, IMHO, means they don't do clustering. There is no cluster-wide locking, software runs on one node at a time, there was a limit of two nodes, and it required a shared disk.
Microsoft Cluster services is designed for one thing: High Availability (little or no down time / load balancing). Beowulf clustering is designed for one thing: Parallel Processing (data analysis / number crunching). They are two different types of clustering. The debate on cost is a waste of time. While Linux is as capable of high availability clustering as Microsoft is, it has little cost. With Microsoft you have to buy a license of Advanced server for each cluster node and then have licenses for each application as well. For cluster aware Microsoft apps that means Enterprise editions. Advanced Server costs in the $4000 range. SQL 2000 Enterprise Edition cost in the range of $11,000 per node. If you are backending a website with a SQL cluster, just for SQL you are looking at around $20,000 per processor . If you are looking for a cluster to be online 24x7 then you go with Microsoft (and pay the additional money for support). If you are looking to predict weather patterns, analyse ocean currents, or predict the lottery, use Red Hat and Beowulf (and pay the additional money for support).
BSODs are an issue for Windows and not Unix / Linux primarily because in Unix video / device drivers don't run in the inner ring of the kernel, and can't bring the whole box to it's knees because of a minor bug in a driver or a hardware failure in (what should be) a seconday I/O device.
Windows NT 3.51 had the video drivers (and most other drivers as well) in the outer ring of the kernel where they couldn't down the whole machine, just certain services. I've seen Win3.51 boxes with horribly buggy video drivers just keep right on running when the video would lock up. NT 4.0 and above aren't the same.
The decision to move the drivers into the inner ring of the kernel is why BSODs are a Windows issue. Blaming the user for not setting up his box just right doesn't solve the real issue, poor OS design.
A real OS (unix/linux/OS390/VMS/even NT3.51) doesn't have these problems.
--- It is not the things we do which we regret the most, but the things which we don't do.
our beowulf team (http://home.cwru.edu/beowulf) used pvm for linux, but pvm runs on everything including the kitchen sink. there are new builds of the win9x/nt versions of pvm out on http://www.csm.ornl.gov/pvm/ . it is a good system, i suggest putting all the doc files in one huge binder. not that bad to work with though...
Windows is a commercial product. It is designed to run well on cheap, affordable computers it runs much better if you throw in an extra little cash. It's specifically not designed to run on obsolete hardware. This also means that no compromises regarding usability and configurability were made to improve performance on old hardware.
IMHO that is a good tradeoff. Running X on a PC with a decent amount of memory and processing power (basically 64MB+, 200Mhz+) is not going to put any significant load on the machine. Similarly, the average windows machine can easily handle both the GUI and server processes. If you are experiencing performance problems with your server processes because of the GUI overhead any responsible sysadmin would upgrade the hardware because getting that close to the performance limit of your hardware is bound to cause you trouble anyway (a minor increase in server load would be enough).
Don't get me wrong, I love linux and have used it on old hardware and found it served my needs perfectly. However, you really need to know your stuff to get it up and running. When it comes to configuring things windows is easy when it can be and just as hard as unix when it needs to be. Basically, for simple server stuff you can get IIS up and running relatively easy. The default setup for apache on the other hand is pretty useable out of the box but as soon as you need to tweak it even slightly you are on your own. For professionals it doesn't matter, they have the time and need to get familiar with whatever they configure. Basically this type of sysadmin is knowledgeable and expensive. You are unlikely to find one in small organizations. Instead you will find loads of inexperienced script kiddies who terrorize their users with major fuckups. If I sound frustrated its because our local sysadmin (linux) just screwed up our mailserver (suse box and some ancient solaris machine) and I'm expecting some important mails. It's not the first time and I'm afraid there's more downtime ahead.
For the casual admin who just needs to get an unfamiliar service up and running with no fuss the windows way of doing things is simply easier. The overhead of a GUI is irrelevant in any business case you can come up with (business cases also include licensing, sysadmin salaries, hw cost, training cost, etc.).
Jilles
The beowulf-specific libraries may be OS-agnostic, but that doesn't mean that all OSes are equal.
How's the networking support? What about system services in general? How much overhead from the OS and mandatoy applications and services? What about libraries for the specific task at hand?
Linux/Unix can be stripped down to almost nothing. But can you kill the GUI of a clustered Windows system?
This isn't a trivial issue - I remember reports of some NT servers that were running near 100% capacity with minimal loads. The problem was traced to a screen saver! Once you know this you can disable screen savers, but what about other unneeded services?
For every complex problem there is an answer that is clear, simple, and wrong. -- H L Mencken
Actually, there are several telnet severs available for Win, plus the wide variety of VNC/SMS, etc control programs.
Unfortunatly, the only way to reduce the overhead is to change OSes. Not an option at times.
--
Evan
"$30 for the One True Ring. $10 each additional ring!" -- JRR "Bob" Tolkien
I've been trying out XP in a VMWare box on my Celeron 450. It not zippy, but a far cry from "barely functions". About the same as KDE, actually, in the same scenario.
"People that quote themselves in their signatures bother me" - athakur999
What? Where? How? I thought that only universities could get "site licenses" (no marginal cost per installation); all commercial organisations have to license each MS product separately (even using Open License or Select License, there's still a per-box cost).
I don't know or care how capable windows is at computational clustering. What I do care about is that windows is designed for several things that you just don't need in a clustering environment. It's designed to be compatible with DOS and older versions of windows - well, you don't need that for custom computational work. It's designed to be user-friendly - but in this sort of project, everyone is going to be technically inclined anyway, so a pretty GUI isn't that important.
Those are the main selling points of windows. But they are *selling* points - you pay for the privelege of using these features. I'm not an anti-microsoft fanatic, but why pay for windows here when Linux or several other free OSes would probably do just as well, with less overhead and cost?
I'm the stranger...posting to
Tell your Microsoft rep, "Oh, I dunno.. Microsoft Linux 2002 has a nice ring to it."
AFAIK, Win32 has a two node limit, but i've heard rumors of 4 and even 8 node support floating around in the form of CA releases. It wouldnt be too wild an idea to see Win32 support 4 nodes in the not too distant future.
Windows is the runt of the clustering litter. Even Novell wipes the floor with Win32 in that arena. The hands down winner is still Beowulf.. you cant beat the flexibility, low-level control, support, documentation, and bang-per-buck ratio that PVM offers.
Cheers,
Bowie J. Poag
Having seen first hand how poorly the following setup ran, I'd say steer clear of Microsoft until they admit that reboots are not normal:
2 x HP Netservers, both dual p2 Xeon, 1gb ram, and a small raid shelf with 8x 9gb disks. Both NT4 installs with the correct patchlevels.
One machine ran oracle, the other IIS, these were clustered so that one would take over the task of the other, should there be a problem.
Problems:
1) Crashing (daily at least)
2) Slow (astonishingly poor, disk defrags once a week helped this)
3) Sometimes one host would freeze, and the other wouldn't actually notice
4) Often a shutdown of one node would move the services across, but upon rejoining the cluster - the node with both services would refuse to give one back.
5) Often, IIS would stop talking, and neither node would actually realise.
The attempted solutions:
1) Replaced CPUs, memory, disks, eventually nodes
2) Reinstalled clustering software, eventually total clean installs of operating system and applications
3) Support from Microsoft, and Oracle, and HP who made the (certified) kit. Oracle+HP both pointed the finger at the OS, Microsoft simply failed to help, when we got any response from them at all.
4) (this helped) I used one of the spare HP9000 servers to monitor them remotely by trying test transactions - it alerted people when they fucked up.
I think the above says it all really. Standard software on correct hardware - it just didn't work properly. Microsoft can stick their clustering "technologies" where the sun don't shine.
--
ALL YOUR BASE ARE BELONG TO US!
I can say that, I went to SIU
My beliefs do not require that you agree with them.
We ran four clusters: two file server and two database. The fileservers were using Microsoft Clusters and the database clusters were using Digital Clusters all of it running on Compaq hardware.
The Digital Clusters were stable enough, but they continually dropped the ball on automatic failover, load balanacing was all done manually. The fileservers were supposed to auto load balance and failover, and did neither. We continually had to reapply permissions and recreate shares on the file servers whenever we would manually failover drives, and BSOD's were an almost daily fact of life.
I would hope that things would be better under Win2k, but the technology under NT was just not there at all.
Wu-Tang Name: Half-Cut Skeleton Get your own Wu-Na
Yes you can build a Windows Beowulf cluster. They work very well. Thomas Sterling just came out with a book describing how: "Beowulf Cluster Computing with Windows" Its a Scientific and Engineering Computing Series book from MIT Press.
Good point, but it's not easy to install, tweak, and generally tinker with things on a cluster of windows boxes that like to hang all the time (despite the stability improvements that showed up in win2k). I suppose it'd be the same with linux if you were tied to some heavyweight windowing system for the vast majority of your system tweaking.
Kinda makes me wonder though. Is it even possible to run a windows machine with no keyboard, graphics card, or mouse plugged in?
Enough material about MS clusters to brag about.
Never underestimate the relief of true separation of Religion and State.
Of course it's possible. It's not *nice*, but it works.
... ;) It's not fast, but since all it does is pass packets through (cable modem) and block incoming packets it doesn't need to be.
For the record, I maintain a headless NT 4.0 web/database server at work for one of my projects (requires disabling the mouse driver to avoid error messages at startup) controlled via PC Anywhere and a headless Win98SE machine at home as my internet gateway (running SyGate NAT 3.0 and SyGate Personal Firewall) controlled via VNC.
Why NT 4.0? Mandated at the time (the main servers are in the US maintained by an external group - we're in Australia with the admin server for the same system).
Why Win98SE? I tried various linux and bsd distributions on the machine, and couldn't get any to work - Pentium 60MHz, SCSI, plain IDE (not ATAPI) so had to install an I/O card to get a CD-ROM to work, old intel ethernet cards, etc. I've configured it to reboot every night so I don't have stability problems
I was actually involved in presales on some hardware for some clustering and which eventually ended up running on UNIX but Microsoft was involved in the early meetings. They sent along the usual contingent of sell-em-anything sales droids but there was an honest engineer with the group who said:
"It's not really accurate to refer to this arrangement of servers as a wolf "pack". A pack is an organized group with a leader working towards a defined goal using a plan with a visible, known structure. These servers just sort of hang out together. It would be more accurate to call them "wolf buddies"."
Microsoft didn't get to put their software on the solution but I did tend to put more credence in what that particular engineer would tell me about the capablities of Microsoft products.
"I can't give you a brain, so I'll give you a diploma" - The Great Oz (blatently stolen sig)
I used to run different shells on Win98. It's pretty much a no brainer to replace explorer.exe all together, despite the FUD spread here. IIRC, there was one called Talisman that let you use HTML to create a desktop and folders. If you like, you could use command.com (or CMD.EXE in NT/2000) as your shell and never have to load explorer. This would be similar to Linux sans X, but would stil allow you to run graphical apps without loading the shell. (On MS, the window manager and GUI tools would still be loaded.)
Really, clustering means so many different things that a question like this just can't be answered.
Even companies like Sun that talk about their 'Clustering technoloy'... are really talking about application specific solutions.
Beowulf? What, you mean using PVM? Windows could do as well as linux, why on earth couldn't it?
Mosix type clusters?
Web clusters that are more of a farm?
It's a silly question.
What's an M$ site licence worth for that site? Ten grand? A hundred grand? That'd almost make it worthwhile suffering through ISO-9002 certification and all of the other bullshit necessary before selling to gummint... we'd sell a Linux site licence for half price and throw in site licences for OpenOffice, AbiWord, SIAG, Apache, Zope, PostgreSQL, MySQL, GIMP, PERL and Python site licences too. (-:
Got time? Spend some of it coding or testing
Getting past what are the wrong tools first: Beowulf is an architecture to do massively parallel computation, so we can eliminate two of the best known HA tools. Microsoft Cluster Service is two or four node high availability, similar to HA Linux's efforts. NLBS is a software form of a hardware load balancer, similar to Cisco Local Directors and only really good for web farms. So what does MS provide to do similar stuff as Beowulf?
COM+ and Queueing Components. AppCenter.
The way it works is this. You write a COM+ component that is transactionally queuing aware. Each component takes a work unit in, processes it, and then sends the result of the transaction to the queueing components for reassembly or re-issue (if a node fails to submit a result, for example, good for checkpointing).
You can use normal Windows 2000 Professional boxes for the worker bees, and use a few Windows 2000 Server boxes to co-ordinate the issuing of jobs and control, and munging the result sets coming back in.
If you need to submit a wide variety of jobs, obviously the COM+ components will be changing regularly, it'd be a good idea to go to AppCenter so that you can treat a bunch of machines as single whole. This allows you to upgrade or deploy an app in a few mouse clicks to literally thousands of machines in a few seconds. AppCenter also has pretty good resource management, something that might be necessary if multiple jobs are running at the same time.
The cool thing is the development environment is really friendly and you can make COM+ components pretty easily and test them locally (for the n=1 case) before deploying to the farm.
There are also specialist MP libraries for the Win32 platform, such as PVM or MPI (WMPI). These have the benefits of re-using the knowledge and API's that users might already be familiar with - one of the biggest thing when a place converts from one supercomputer to another is rejigging and reoptimizing the code for the new architecture.
Andrew van der Stock
neat...but no where near as practical as the yellow dog linux solution...with drive bay sized nodes that just slap right in...if you want to string a few mac towers together I'm sure appleseed is great...but if you are going to invest the money required to to buy decent g4 towers why not take the same money and invest is a node based PPC system where the hardware and software was actually designed for high performance computing.
Apple hardware isn't cheap...and it makes no sense to make an investment in stylish apple towers when what you need is raw cpu cycles...you get much more computing for the money if you buy into ydl's little briq nodes....
-jef
Can you imagine a Beowolf cluster of these Beowolf clus...
Oh, never mind.
D'ya think this might be a response to Samba's CAL server ability?
BTW, don't knock savings on CALs. Get 1000 students onto Exchange and you're staring down the throat of at least $USD30,000 in CALs alone. Get 1000 students into PostFix and you're staring down the throat of, well, no licencing costs at all. And you have capacity on the same hardware to serve roughly four times as many mailboxes...
Got time? Spend some of it coding or testing
Actually, it does loosen the screw. The rellies on the farm use a hammer quite effectively as a screwdriver (both ways) and spanner where appropriate. It just doesn't do the screw in question much good...
I've used a hammer myself to gut a dead hard drive for the magnets, when I didn't have a small enough star driver. I just flattened the top of the bolt out to tinfoil thickness and pulled it straight through the metal cover. The technique with screws is different, some light taps can loosen them in their substrate (typically wood or sheetmetal) enough to winkle out by hand. Using somer CRC/WD40 often helps as well.
Got time? Spend some of it coding or testing
It's up to the vendor to prove that it works and that it will be good for you. If they can't show you anything better than what you already have then ignore them, even if they are Microsoft. If they can show you something better, then look for holes and unanticipated costs. Also look at the companies with experience at that end of the market, and not just a big PC company starting to ease into the top end (which is really what they are - hugely successful, but a PC company). It may be a good solution, just get them to prove it.
The first thing that sprang to mind when I read this was, "Microsoft cannot compete, since Windows depends on the GUI."
Remember that one of Microsofts contentions in the anti-trust trial is that they cannot unbundle Internet Explorer from Windows, that the system is so interdependent that no elements can be left out and still function.
So they cannot compete on price, since all other things being equal a Windows machine must have a video graphics card.
They cannot compete on performance, since all other things benig equal a Windows machine must spend resources on storing and running the GUI.
Yesterday, I was showing a very happy WindowsXP owner (who also happens to be a somewhat savvy computer consultant with Unix and Linux experience) the beauty of Debian's apt and dselect packages. He was so happy with the granularity of not installing anything that he doesn't want, that I gave him my Debian 2.2r4 CD. (I'm running Woody anyway)
Bob-
The Ludwig von Mises Institute. The reasoning individuals economics
I can't speak to the quality of a Linux Beowulf cluster. I did, however, recently setup and install an MS Win2K two server cluster for a large manufacturer.
The hardware:The Software:
The purpose for using a clustered setup was for failover assurance. The active cluster controlled the manufacturing process, and if the cluster went down, the whole plant went down. We settled on a two node cluster. We later discovered that two nodes is the maximum that microsoft "recommends" (read supports). The cluster was to act as the gateway between the plant IT network, and the device network of PLCs for the machinery. The cluster would serve progress reports to the IT network, while controlling and collecting data from the device network. In the case of a failure on one node, the second node was to pick up the load.
After our initial research and development on the project, we became seriously disenchanted with MS clustering. No synchronization, no active replication, no load balancing.... about all it could do is fail gracefully from one node to the other. The three NICs on each node were a real nightmare. First NIC was a heartbeat. Second NIC to the plant IT network. Third NIC to the plant device network. We routinely encountered routing table loops through the heartbeat NIC, causing VERY long delays in connections.
The RAID acted as a quorum drive between the two nodes. The active node had control of the quorum, and the MSDE database was installed there. When a failover occurred, the last transactions committed were in the database. Unfortunately, any data from a transaction in progress was completely lost. The nodes often got confused over who owned the quorum, sometimes both fighting for it, other times both dropping it.
Fortunately, this system is only a prototype, and never controlled anything more important than a machine simulator and an isolated test network. I would highly discourage anyone considering using MS clustering. If you are hosting a web site, you may get some use from it, but if you are looking to perform distributed transaction processing, constant uptime, and at the same time provide a robust system, avoid MS clustering.
a beowulf cluster has three main components, not including the application running on it.
1. A big pile of commodity computing hardware and a fast network interconnecting it.
2. An implementation of distributed computing API that handles things like message passing between nodes.
3. Utilities to manage the cluster as if it were a single machine (like starting a copy of the application on each node, etc)
Looking over this list, windows can run on the same hardware as linux.
There are windows implementations of the Parallel Processing APIs typically used by applications on linux beowulf clusters. Are they as efficient? I don't know. Maybe someone else does.
That leaves 3. The tools for managing the clustered application as if it were running on a single machine. I don't know if these exist for windows now, but I see know reason such tools couldn't be created with relative ease. Windows already has extensive infrastructure for remote monitoring and management.
Asking slashdot about Windows Clustering is like asking the Pope about the benefits of Satanic worshipping...
I have no problem with your religion until you decide it's reason to deprive others of the truth.
I didn't say that any Linux-based alternative is better or cheaper than WLBS, although I think that NAT-based load-balancing solutions like the ones offered by Linux offer the benefit of better troubleshoot-friendlyiness than WLBS. I agree that they scale worse than WLBS, but the fact is largely irrelevant for loads up to - I think - a significant chunk of a megabit of served bandwidth - which for a typical non-static windows-based site means at least a few dozens of servers, at which point the cost of a proper hardware load balancing solution becomes more than acceptable.
Also, WLBS works by tricking switches and forcing them to flood all servers with the data packets. This limits your total served bandwidth severely, doesn't it? It MIGHT be very light on the servers, but then if you have N clusters on a single LAN Segment, maybe 100Mbps switched, each cluster gets on average 1/N of 100Mbps. Sure, it can be worked around by using L3, but still you don't scale past 60% of 100Mbps (the remaingin 40% is lost due to to ethernet broadcast collisions - a conservative estimate).
About WLBS working with everything, try to tell that to an Ethernet-over-ATM switch. The trickery there won't work.
About the instabilities: we were running Apache 1.3.something for NT4. It MIGHT have been the custom-code, or it might have been not. A fact is, since they've been moved to an external load balancer those exact same serves were way less crash-prone.
About using what the business needs: I agree 100% with you. But the business also needs something that works. If you need 24/7 coverage to reboot servers as they stop working, it's pretty hard to say the words "TCO" and not start having fits.
As I said, WLBS is gone and is not missed.
Whoa!! Could you imagine a Beowulf cluster of Beowulf cluster jokes?!! Man, that would RULE!
pr0n - keeping monitor glass spotless since 1981.
1. "Windows clustering" AFAIK, only includes load-balancing and some transactions processing, designed for web servers primarily. No one uses it though because application-specific designs are usually better. Cluster management also exists, but I don't think, it's helpful in general.
2. There probably won't be much trouble to put MPI or PVM on a bunch of Windows boxes, and run something clustered. It's however unlikely that this thing will be easy to manage, and certainly the performance is likely to suck because Windows networking was "optimized" in various ways that make sense in a, say, web or mail server, but offer no help otherwise.
3. Windows has no equivalent of MOSIX or other advanced clustering architectures, and messiness of its design doesn't give much hope for it.
Contrary to the popular belief, there indeed is no God.
Allow me to one-up myself, then:
ALL YOUR BEOWULF ARE BELONG TO CLUSTER!
pr0n - keeping monitor glass spotless since 1981.
Maybe finger was the trial version of Exchange? That would explain why the code name for the full version got censored...
Got time? Spend some of it coding or testing
Pedantic aren't you? Of course what I should have said is that you are required to run a GUI of some sort (unless there's a lightweight console-only shell) on Windows, you aren't on Linux. That's how you can still install a RedHat server on a 486, while anything beyond Windows NT will choke (and NT would be a serious stretch).