Slashdot Mirror
White Box, Or Big Names for Lower-End Servers?
LazloToth asks: "Those of us who manage small- to medium-size networks face the decision all the time: for the run-of-the-mill web, print, or storage server running on i386 architecture, should we buy HP or Dell, for example, or build it ourselves from commodity hardware and save some bucks up front? In my operation of fewer than 50 servers, one will see a mix of the two. For servers that take more abuse, I tend to buy the proprietary stuff. But not always. I wonder what experiences other admins and managers have had with do-it-yourself servers in a production environment, and whether they feel that white-box servers perform as well - - and last as long - - as anything else? What is the mix in your network of big-names to no-names?"
there are two sides the issue here
big name - warranty (saving your ass)
white box - if you build it yourself you know what's in there. It's cheaper. But you don't have a warranty.
Show this to your friends and family that don't know what a real hacker is
I bought a noname 2U server at $600 CDN for the company which gave us alotta grief... modems would never work in its PCI slots. So we decided to always go proprietary. All our big servers are ibm xseries.. and we buy ibm xseries 206 ($600 USD) for cheaper stuff. We never go Dell on servers.
I know you can build a superior system thats whitebox. MDG sells machines for cheap with Intel motherboards. You can buy Tyan mobos for whitebox systems.
However keep support in mind. Everytime something breaks on IBM xseries servers, we call tech support. In 4 hours of calling the replacement part arrives, and the techie arrives the same or next day and replaces the part no questions asked. Sure we've had lots of trouble on our tape drives etc, but it gets replaced painlessly, no driver changes, and no financial hits.
Another benefit of name brands is that you can say youve worked on so and so servers in your resume. Smart employers wouldnt or shouldnt count that, but you do see people asking for MCSE and proliant servers, etc. Its even more specific when you get into UNIX... they'll only accept that brand of unix.
"Give orange me give eat orange me eat orange give me eat orange give me you." -Nim Chimpsky
Lights out management usually works better on IBM, HP or Dell systems. Also, building and fixing machines is a pain and gets time consuming & expensive, particularly if you get a bad batch of drives or motherboards that requires alot of fixing.
If you are running < 10-15 machines, I can see cost savings in going whitebox. But if you are tight on staff and runnings lots of machines, buying name-brand kit is cheaper in the long run.
Conformity is the jailer of freedom and enemy of growth. -JFK
The contents of the box are pretty generic for most purposes. Motherboard from Asus/Intel/..., BIOS from Phoenix/Award/..., Processor from Intel/AMD/Motorola/...
What really makes a difference is the vendor. I have a local guy who I can call and ask for recommendations and advice. If I tell him I want a Dual Opteron with 12 gig RAM, mirrored 74 GB hot-swap drives, dual hot-swap PS and a rack-mount case of my choosing he personally delivers it a couple days later.
Drive in my raid-array dies? He brings by a replacement the following day.
Oh, and the only number he gives me is his cell phone. And he answers it. Always.
With the exception of some specialized telephony equipment (actually a different white-box vendor specializing in that market - Dell et. al. wouldn't have a clue about this stuff), he is always my first call.
I've been using him for years. When the company he worked for ceased operations he started his own and service has remained outstanding.
I guarantee that nobody who uses the "name brand" machines can come anywhere close to the responsiveness and support that I get from my local vendor.
~~~~~~~
"You are not remembered for doing what is expected of you." - Atul Chitnis
We went the white-box route on our first compute cluster, which were then converted to desktops later. Decent machines, but the power-supplies weren't up to the 24x7 operation and tended to eventually have the fans sieze up, causing the ps to overheat. Eventually other components showed that they could have been better, and we cannibalized some machines to keep others running. They were replaced by HP and IBM boxes under 3-year, next-day, service contracts.
The advantage of calling IBM, HP, or even Dell, is not simply the service contract (though your time is worth something), or the fact that their QC is superior to wherever you're getting your parts from, but that they have real engineers, who worry about such issues as optimizing air-flow, choosing proper fan-sizes, etc. Take apart an IBM xSeries 345 some time, then try to decide if you could actually buy parts to build a machine like that, for less than just calling IBM.
White-box systems may have once made sense, ( I remember a 386/40 AMD-based system that I wrote my thesis on that was still running when I came to visit years later), but with modern components, heat-loads, etc, it pays to invest in properly engineered hardware, backed up by a company willing to service it on short notice. WB hardware may still make sense for desktops, if your environment keeps the data in non-local storage, so that a new desktop can be dropped, booted, and put into production immediately. Never with servers.
We adopted an informal, simple, but effective policy: Do not buy any machine that doesn't come with a three-year warranty, or hard-drive that doesn't come with a five year warranty.
the more accurate the calculations became, the more the concepts tended to vanish into thin air. R. S. Mulliken
With the low price of low end servers you aren't going to save a lot of money going with the low end box.
Consider this, you buy, or build a white box. You'll end up with very short warranties provided by different companies, very much a pain in the arse. You may save $200. Now think of how much a 50k/year sysadmin makes per hour (roughly $25 if I did calc right)
So, $200/25 = 8 hours...
Now you've got $200 which is equivelent to 8 hours. Are you going to spend more times on a whitebox than a dell? I would say so. ESPECIALLY if you're building yourself. Consider extra time spent finding parts. Extra time putting it together. Then when things fail you have to round up the warranties for individual parts. Probably your warranties won't be as good as what dell provides. And then repair time. I know as a sysadmin we tend to repair ourselves anyway, but consider it may be something you WOULD let an on-site tech repair because you're busy...
In a home situation, I'd say build your own. When you're off the clock, your time is free. but at work, when time IS money, buy the named stuff.
BTW, my numbers are BS. Play with your own, I think you'll draw the same conclusions.
Do not meddle in the affairs of sysadmins, for they are subtle, and quick to anger.
Just because it has a "name brand" like Dell doesn't mean it's any good. I've been highly unimpressed with their hardware. Just this last week I saw a brand new Dell server in which the hard drives had a plastic cover that made air flow impossible. At another place in the case, there was a fan whose intake was a closed plastic area. Hot hard drives and placebo fans don't improve server quality.
Whether you do it yourself or buy a name-brand system, make sure that the case is well-designed and that the components are high quality.
I support 2 Win Server 2003 boxes, 2 Redhat 9.0 boxes and about 10 desktops running XP pro. All built from commodity hardware. I use Abit motherboards, Pentium 4's or Celerons, Intel NIC's and ATI Graphics. I've had *zero* problems after the initial startup issues of a bad RAM module and a bad ATI board. I also suggest Toshiba DVD/CD writers.
"Eve of Destruction", it's not just for old hippies anymore...
If it's your machine, buy it, build it, make it exactly what you want for a good price. If it's for a company and you may or may not be working for when it needs support I'd say buy from someone where it has a model number. It makes it a lot easier to know what it is.
If you are comfortable building *and supporting* the machines through their life and there really isn't anything out there that's pre-built and has the right price and features it might be worth building. There are times when the market falls behind what is capable with current technology and nobody is configuring machines in a way that's both possible and useful. I'd only build your own if you fall into this category. Otherwise, Dell can often build stuff cheaper than you can. There are exceptions though. I find them incapable of selling a good storage subsystem for a reasonable price. It's been a while but the last time I tried to find something with a good 4 drive SATA raid5 configuration I found nothing reasonable.
As far as "white" box goes, I never skimp on the box. A lot of the stability of a machine is dependant on the power supply and cooling so its a good idea to go with good stuff. The other key to a good system is a well designed and tested motherboard. Gamer and desktop motherboards are getting a lot better than they used to be but are still a bit on the flakey side (new ones are still arguably better than old server class motherboards used to be though). Often, when you price things out using high quality components you see why the server class machines end up costing so much and the benefit of building your own goes away.
set softtabstop=4 shiftwidth=4 expandtab nocp worlddomination
In terms of service.
IBM's 4 hour turn around on parts and service can save your arse if you have to have all your eggs in one basket. Even a failed hard drive in an array is a looming disaster you'd like to fix now, not next business day. So even at 3am on Saturday - you can have a drive delivered so you can have your array back to 100% by the next business day.
If you go with whitebox - be sure and build in redundancy so you can lose one, or take it down for extended repairs. Because the money you're saving on going generic you're also losing some of the benefits to using big iron. Diagnostic tools, troubleshooting support, features like LightPath show you the failed components fast, raid backplanes, firmware, and software that is consistent, redundant hardware like PSUs, drives, memory failover, etc etc... There are a lot of benefits, but it costs money that sometimes you can put to better use.
The hard part is deciding which servers are going to put your ass in a sling when they go down for an extended period of time. And to most end users - 30 seconds is way to long, slacker!
After I lost my second finger in the sharp guts of a whitebox system while trying to fix it (again), I decided to go with brand-name and I never looked back.
I work in a fairly large datacenter, where I help support many of our colocated customer's equipment. Some of which we sell and maintain, some of which they purchase and colocate. I've seen a good mix of generic servers that were custom built because there was no pre-built options available. But when it comes to support - there are few options when things go down.
Motherboard company blames ram, ram company blames raid card manufacturer, raid card says it's a bad firmware version of the motherboard... two hours later, server is still down. Who's going to let us swap out a motherboard just to see if it works?
I don't see a price advantage to whitebox servers compared to modern server hardware from the big names. Anyone who's just looking at the price tag is fooling themselves.
Dell's hardware is unimpressive, I'll give a nod to the previous responder who mentioned that. And storage subsystems are still insane. Even with the evolution of SATA for slower mass storage, cost/MB is still too high with these subsystems.
Beware of a name brand's inexpensive servers. Some of the rock bottom units are cheap, but they lack some of the basics like raid, hot swap drives, expandability... On the bright side, even if you go with these cheap units, you'll still have service and support from a major player.
The main difference you get with HP, IBM, Sun vs the rest is the manageability of the hardware.
A generic box fails, or has intermittant failure, and sometimes you are scratching your head figuring out what is wrong. The better designed gear will tell you that "Dimm 2 has been throwing ECC errors for the past couple of days". Gives you a place to look. In the generic box, you are replacing all the RAM sticks.
I don't see a whole lot of difference between a Dell and a whitebox.
For the Big, Important Servers-- customer-facing web servers, product db backend, major fileservers, etc-- then IBM, HP, NetApp for fileservers, etc seem to be the way to go. Use somebody who's built a name for themselves in the enterprise by service, not marketing; Dell still doesn't know how to support those needs.
But for intermediate servers-- internal web servers, testing boxes, etc-- you can go with a smaller company. It's still worth going with a company, rather than DIY: the company deals with fixing servers every day, has the parts on hand, etc. Your organization may have great people, but the guy who is constantly building servers for a living is going to beat you on service.
The smaller companies, like OffMyServer (blatant plug for a company who's done well by my employer), can meet your needs without breaking the bank. We have dozens of servers in my department alone, and we just couldn't afford to put a big HP contract on each of them.
ObDisclaimer: Speaking for myself, not my employer, my own opinions. Not affiliated with any of the above companies that I know of, other than that my company buys from all of them.
I have had good experiences with HP, IBM, and Sun support. All have sent out technicians to fix problems promptly. Dell support is alright, but you need to twist their arm some time.
For rackmount gear go with a name brand. I have had nothing but trouble with generic white box rackmount gear. Recently a stack of 20 antec cases was 1/4" too high to fit in the industry standard rack.
For non rackmount servers I will go with HP/IBM/Sun if I want SCSI or similar server features. For really low end stuff I might go with white box but only if the hardware budget is an issue, or if I need a specialty box with specific hardware. If I go with a white box I always use higher end components so their isn't much of a price difference anyway.
The biggest issue I have had with white box machines is that the hardware was not designed to run 24/7 and it fails. Despite what the tweakers think most white box server cases have poor heat management. Adding more fans is not the solution when the harddrives sit in a dead zone of low air movement.
And again the support from HP, IBM, and Sun is really nice.
One of the questions that you should answer for yourself is, "If this server is down, how much time will it take the boss to get pissed?" If the answer is less than one day then get a name brand and a service contract that guarantees a fix, but from the sounds of things, the answer might be more along the lines of multiple days. So as long as you can make it work then white box might be a better idea in the long run, if you can handle 3 days minimum for a part replacement.
Just my thoughts on the issue.
I have a server room filled with HP servers. We lease them for 3 years and then send them back. I have ZERO problems with the software and I get parts (usually drives) replaced in less than 24 hours.
These servers run 24/7 for the 3 years we have them and give us no problems at all.
I buy second hand named brand servers.
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
You take the four year old Dell Optiplexes within your organization that otherwise just go to a salvage aucton. You install Linux or a BSD OS on them.
'Big Name' at lower-than-white-box prices. Voila!
resigned
I bought a few white boxes from a noname vendor. Service was horrible (this is my experience only, yours may vary), and ended up spending more than I saved in time. It took me 4 weeks to get a new HD for the machine. With IBM or Dell, it is delivered via Next Day Air at a minumum, usually a courier within 4 hours. The time you spend screwing around with the cheap servers will quickly exceed the money you saved. There is also the fact that the cheapies never quite fit in the rack right, they don't have any cable management equipment, and the cases are usually so cheap, they will cut you every time you open them.
I think the general consenus around here would be to go with Dells, IBMs, etc.
We use Dells. Partly, because they fully support the IPMI protocol. It's really nice to be able to remotely control the machines, see trend statistics, etc. using IPMI.
Don't think that a small group of dedicated individuals can't change the world. It's the only thing that ever has.
The price point of a server includes far more than the hardware in the box. I'd say the hardware itself is the least costly part of the package. Others have already mentioned the warranty that comes with an HP or Dell box. As important though is the software and support that comes with the box. For example, having the HP management agents installed on a server is a necessity if you need to support more than a few boxes. HP's Integrated Lights Out feature is a huge plus, especially in remote environments (remote power switch is great!). HP has knowledge bases available on their site, new drivers and firmware are always available, documentation can easily be downloaded, etc. Having been a white box vendor for years, trust me when I say a white box can't hold a candle to what HP can deliver. I've never used Dell, but I imagine they are comparable by now.
Print server: If something craps out, and it takes me a couple days to get a replacement parts, nobody is going to have a fit. I use a white box, or salvage whatever we've got. Saves money.
Production database: Something dies. Vendor technician is there in 3 hours with a replacement part, and you're back up in 4 hours.
For random office machines, we use whatever we've got around, or buy something cheap. For production, its all IBM or Sun.
For developer laptops, we use IBM. Prices are high. Specs are low. But they're solid. (I wish I could get Apple Powerbooks, but IBM doesn't have a db2 client for osX or linux on powerpc.)
The masses are the crack whores of religion.
We buy hundreds of white 1U pizzaboxen from SiliconMechanics.com every month (not only are they white, but they are also blank -- we net boot debian GNU/linux onto them ourselves). SM has an excellent record for replacing broken parts, although we're never in an emergency when something breaks since we deploy backup hardware for everything. If something breaks we can switch to the current backup, start converting a spare machine to be the new backup, and then take our time getting the broken hardware fixed, its all under warranty.
All of our vanilla services: mail, web, and even database are on white boxen from SM. We have some black box stuff for heavy mass storage.
Religion is poison to rationality, and we lose sight of that at our own peril. -- Lurker2288
Besides, by depending on this guy, you've created a one-man point of failure. What happens when this guy gets sick or goes on vacation? Where's your immediate response then?
Even if he never gets sick or takes time off, he's not going to be able to sustain this level of service. His own good reputation will work against him. He's obviously one of those people who has to do everything himself. He's probably not very good at delegating or training, so he's never going to be able to scale up his operation. So unless he starts turning away business and dropping customers when they get too big for him to handle, he's going to get in out of his depth.
If I were in your shoes, I'd want my hardware needs met by a solid organization, once I could count on not just now, but years from now. And that has to do with people, not with where the boxes are assembled.
If you have enough warm spares ( and staff ), then generic servers should be ok to get by with. The fact you have no service agreements should be transparent to the users.
If you cant afford the extras, its hard to beat on-site service you can get from the big names.
---- Booth was a patriot ----
Using both Dell and IBM I can say that I have experienced the opposite when it comes to support from IBM and Dell. Recently we had a hardware issue with an XSeries server which required a IBM tech to replace on-site. It took three calls to get the incident logged and then the ticket was passed between three different techs before we found out they were busy with other problems and never showed up that day. The ticket was assigned to a different tech the next day who could not make it onsite and passed it to another tech. The other tech finally showed up at the end of the business day to the datacenter and did not have *any* identification to prove who he was to get into the datacenter. Apparently he lost his wallet somewhere and did not think it would be a big deal as he could vouch for himself. He just left and gave the ticket to someone else who showed up on the third day; this being one-day support.
Combining this with the fact we have had a 30% hardware failure in the last batch of XSeries servers we won't be using IBM anytime in the future.
Contrast this with Dell in which they have always been ontime, had the appropriate parts (yes, IBM techs had a problem with this once) and Dell's hardware has outperformed IBM's in terms of failure rate our next big batch of servers is coming straight from Dell.
Quality Hosting e3 Servers
Stealing from a post I wrote earlier this year on a similar subject, I will agree with the folks pushing for name brand hardware instead of hand-building each machine:
Resist the urge to buy/build one-off servers because they are cheap. The $300 one-off computer that some kid built in his garage is going to cost you way more than the difference it would have cost going with a single standardized platform - over the life of the machine.
One person can maintain 300 machines if they are all exact clones of each other. If every machine is unique it would take you 5-6 people keeping the same network fully operational. At $65k apiece fully loaded salary that's a third of a million dollars more per year to support the same 300 machines. At four year turnover on computers, you are talking about an EXTRA $4,000 per computer to save $200 total on purchase price.
Glonoinha the MebiByte Slayer
Decent machines, but the power-supplies weren't up to the 24x7 operation and tended to eventually have the fans sieze up, causing the ps to overheat.
Oh yeah, big time Achilles Heel of the generic PC, assuming name brand mobo and stuff.
It's just impossible to get a good power supply in a generic PC. ("Good" means built with decent quality components, like the Astecs and Lambdas you'll find in proprietary systems. It does not mean "Comes with a ThermalTake Fan and is the choice of 14-year-olds and overclockers!".)
My best success was based on a simple formula: the power-to-weight ratio. Buy the heaviest supply marked with a given advertised wattage rating.
Then, for server use, step 2 is to open up the supply and replace all the made-in-Bangladesh-or-Taiwan-or-China electrolytic capacitors with Spragues or Nichicons rated AT LEAST 1.5x the voltage ratings of the capacitors which were in there. And then out comes the no-name 12V fan, only to be replaced with a (loud! expensive! moves a hell of a lot of air! lasts forever!) Comair Rotron 120V fan running directly off the power line. Also gives you a chance to fix the *many* cold solder joints you're likely to find in commodity power supplies. All told, usually under an hour per supply, with the new fan often costing more than the supply!
Since I started doing this, I haven't had a single failure of one of my white-box server supplies.
Fire and Meat. Yummy.
But you need to pay extra for it. It's called Premiere Enterprise Support Services, and it's top-notch. Direct calls into tier3 engineers, a "SWAT Team" sent out if the call can't be resolved quickly, no questions asked parts replacement, etc. They finally "get it" where enterprise servers are concerned. I've worked in both large Dell and large HP/Compaq datacenters (100's of servers) and found Dell to be more responsive and adept at services than Compaq (which is odd because HP by itself had great service).
Sun x2100 - $675 for a barebones system, just use whichever SATA drive you prefer, and brand name PC3200 DDR400 RAM (I use kingston mostly), voila an Opteron based box with 2 hard drives (there's an onboard RAID card as well), dual gigabit ethernet (broadcom, very nice), up to 4 gigs of ram (4 slots), lights out management, service light, great airflow, serial port that allows BIOS access as well as OS access (most whiteboxes won't do that). PCI-E slot for a raid card/fiberchannel/ethernet/etc. Oh and that $675 includes shipping and handling (which makes a big difference, shipping on a 1U is typically 100$ US with insurance/etc). These machines run Windows, Solaris, Linux (certified for Red Hat and SuSE) and OpenBSD (undeadly.org is running on a Sun x2100 as we speak).
http://undeadly.org/cgi?action=article&sid=2005111 2002121&pid=2
proprietary hard disks in HP/Compaq, IBM etc? absolute shite. show me a single iota of evidence this is true in say, a proliant. they're just commodity scsi disks in a vendor-specific sled.
you no longer have warranty support. this means should you need spares, you're going to struggle to get them quickly. it can make a lot of sense, but in other ways you're getting the worst of both worlds: the lack of support of white boxes, and the expense of running named servers.
it can work, but you need to weigh up the pros and cons - and be sure that you're qualified to do so.
my money's on new, named boxes, and replace them every 3 years unless you're really small or really cash-poor.
"run-of-the-mill web, print, or storage server running on i386 architecture"
You don't have any old PII or PIIIs laying around?
We often retire old desktops to print or small web servers at my office. We usually have more of them than we know what to do with. Now for storage that is a different story. If a print server goes down you probably didn't loose anything. For storage I would want a good white box or a name box.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
We use Optiplex GX110 - 500MHz, 256MB RAM. They're cheap, really easy to get hold of second-hand and Linux runs fine on them. They're so cheap we just buy twice as many as we need and use the others for spares.
You work for a hospital?
Try finding a ready-built server with an ARM or PWRficient chip in it... Whitebox has the nice full featureset, so its a good platform for running a server off. That said, I may still buy RHEL because the support is excellent. All the same, it's not worth subscribing for up2date, i find yum a better alternative.
~HTP~ Hug that tux
This has been beat to death before. I could recall this one, can probably find you even more hits.
1 45234&tid=98
d =11301686
The thread:
http://ask.slashdot.org/article.pl?sid=05/01/09/0
My take:
http://ask.slashdot.org/comments.pl?sid=135424&ci
Quote:
I wonder what experiences other admins and managers have had with do-it-yourself servers in a production environment, and whether they feel that white-box servers perform as well - - and last as long - - as anything else?
99% of the time as an admin, and person-with-their-ass on the line for uptime of servers I could choose to build, I've only seen white boxes, and home growns peter out and fizzle faster than boxes built by vendors I have a service contract with.
A prior arguement someone made with me was " its the same parts ". In some cases, they are. However, local Joe's QA lab is a fold out table with static bags duct taped to it.
If it was worth my boss telling me to build it, assigning a budget and delivery date, it was worth a request for quote from server vendors.
Common question to boss: What kind of service do you want ? 2hr parts delivery, or whenever (vendor) gets around to it, or do I have to hope has what we need at 0430 when a power supply goes fizzle ?
Common answer: We use (XXXY)level of service for everything else, do the same for this one.
Whiteboxes and home-growns dont have a service organization built to keep your unit in business.
I contacted the EBay seller and he confirmed, that he tested the disks in an IBM system -- I had to send them back to him :-(
I guess, IBM does this to justify its markup. But it does happen, so eat your shite back.
In Soviet Washington the swamp drains you.
brand name servers is to vary the maintenance contracts. I have an Oracle production, test and development environment. The production environment gets gold service, even though it's a RAC cluster because if something breaks I want it fixed now! The development and test environments have next business day. If we lose a whole RAC node in test we have to wait at the most three days (box goes down on Friday before a three day weekend) before we get it fixed, which we can live with. Going to NBD from gold service on Dells saves you about $1,500 per system, which makes up for part of the cost difference between a Dell and a white box server and I have the peace of mind of knowing that I have the same hardware across my environment so that rolling something from the test environment to the production environment will just work because it's the same hardware.
cheap labor conservatives - they want to keep you hungry enough to be thankful for minimum wage.
The IRS