Slashdot Mirror
A Look At One of Blizzard's Retired World of Warcraft Servers
MojoKid writes "At last count, Activision Blizzard pegged the number of World of Warcraft subscribers at 10.2 million. It takes a massive amount of gear to host all the different game worlds, or realms, as they're referred to. Each realm is hosted on its own server, and in late 2011, Activision Blizzard began auctioning off retired server blades from the days of yore to benefit the St. Jude Children's Research Hospital. They sold around 2,000 retired Hewlett-Packard p-Class server blades on eBay and donated 100 percent of the proceeds (minus auction expenses) to the St. Jude Children's Research Hospital, which seeks to advance the treatment and prevention of catastrophic diseases in children. This article has a look at one of those retired server blades."
For the Horde, I mean, FOR THE CHILDREN!
With shipping, which was almost as much as the server itself, I paid $243.50 for this showpiece.
Hmmm $100 or so to ship? Someone's padding that expense line. I would not flinch at $25 to $50 but this smells of those ebay auctions where its $0.01 for the product and $50 to ship.
"Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
It looks like any other blade, once you ignore the marketing decals put on it.
For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
vmware still has issues with all the I/O going through the hypervisor. the blades have local storage for the OS
there are still applications like Cognos and others that say that if you run them in vmware then use a separate physical server due to the I/O demands and the fact that you have to specially code around the oversubscribing feature of vmware
Um, where do you think those Xeons and Opterons are installed? In individual towers? 1U servers are basically the same as blades except you have a lot more smaller redundant parts (power supplies, fans, etc...).
Plus, you are griping about hardware that has been retired.
I read the internet for the articles.
From the slide show/article it says the drives were removed before hand to prevent customer info from being leaked.
I'm wondering why these had hard drives with data on them at all. Wouldn't the data be on a SAN on the backend? Kind of defeats the purpose of a blade in the first place, seeing you want to be able to replace it quick if something goes wrong.
In fact, if there are using the local drives, they better be sure to remove the RAID controller, as these might have info left in the cache as they are battery backed up.
My sense is that 'blade' as in "Wow, look how many basically-just-a-1U-with-the-economics-of-a-laptop we can cram into a proprietary cage that costs $15k empty!" isn't as trendy as it used to be; but that some of the cuter setups that offer integrated switching, dynamic allocation of a pool of disks to individual blades, and other functions that help save on switches, cabling, SAN architecture, and so on were still in a slightly tense state: On the one hand, they had the potential to be more cost effective that the discrete stuff, for certain applications(because they genuinely saved on interconnect silicon, cables, and various overprovisioned-because-every-1U-needs-to-have-one components); but vendors were having a terribly difficult time resisting the temptation to use the fact that you had to purchase all the bits from them to start gouging and slacking off on standardized management interfaces.
Hipster IT admins, ASSEMBLE!
"Blades are so mainstream. People in the know use a CPU with buttloads of cores. I'd tell you what brand, but it wouldn't matter, you've never heard of it."
"Sure, 10-speeds save physical space, but they add complexity, cost, points of failure, and the heat they generate is the same. That's why I prefer fixies."
"Why mess with blades? You can't even put a bird on them."
The article writer doesn't mention the specs of the blade, isn't interested in knowing if it works and thinks its ugly?! He has no interest in server tech or playing wow. Why waste our time linking to this article?
Oh man, where was the news story when these were still for sale?! $200 for a blade server doesn't sound bad, but then you look at the work they did with the paneling and the plaque and this thing looks like a pretty sweet piece. Practically belongs in a museum! $200 seems like a steal.
:(
I want one
GCS/MU/P d- s:- a-- C++++$ UL++ P+ L++ E+ W++ N o K- w--- O M+ V- PS+++ PE Y+ PGP t+ 5- X R++ tv+ b++ DI++ D++ G+ e++ h-
The main saving grace of the humble 1U is that it doesn't have a vendor who has you by the balls for the next 14-ish systems you buy, along with a variety of option cards and things. Your basic rack doesn't provide much in the way of amenities, leading to lots of messy duplication of 40mm jet-fans and PSUs and a cable mess; but it just doesn't have the lock-in of a physically and logically proprietary cardcage...
So far, the blade guys have had a difficult time not pocketing as much of the extra efficiency value as they can, while the commodity 1U knife-fight is wasteful; but it is rather harder for your vendor to achieve market power over you.
Um, where do you think those Xeons and Opterons are installed? In individual towers? 1U servers are basically the same as blades except you have a lot more smaller redundant parts (power supplies, fans, etc...).
Plus, you are griping about hardware that has been retired.
1U servers are basically not the same as blades, lol.
It only makes sense to employ people if you have a job for them to do. If Blizzard had nothing useful for them to do, keeping them around to twiddle their thumbs doesn't make much sense.
I get the feeling their backend design wasn't the best. For years they took their servers down every single week for a massive 6-8 hour maintenance period. This wasn't for updates, this was just regular. Patches took forEVER to happen. It clearly wasn't something like "Take things down, roll out new code, run checks, bring it online." Given that some things would only affect particular realms it was pretty clear they were doing things like running series of scripts and commands to upgrade things, and the process shad trouble in certain configurations and so on.
So it wouldn't surprise me if they did things like store data on the blades themselves and so on. I can't say for sure, since Blizzard has been secretive to the point of paranoia about how things work on the back end, but my experience with the game leads me to believe they did not have a particularly good backend setup.
Blades are all about density. If I can squeeze 10 dual-host blades in a 7U enclosure, that's 13U saved vs 20 1U servers. Add the fact that many modern blade enclosures integrate modular switches, and you can squeeze 120 hosts per rack, instead of just 38-40. The hardware cost difference is negligible, since you're buying one set of redundant power supplies to power all 10 blades. The enclosure itself is costly, but the blades aren't much pricier than a comparable server board.
If you're deploying lots of them like Blizzard, choosing blades means you only need 1/3rd of the floor space, 1/3rd the shipping cost, 1/3rd the installation labour, which represents a huge chunk of change when you're colocating at top-tier datacenters all around the world.
Blades may not make sense for everyone, but don't write them off just because your needs are satisfied by simpler solutions. Virtualization is a great tool that offers tremendous flexibility and reduced costs, but it is not a magic bullet to solve every problem. It excels at handling small jobs, and fails hard with large ones. For example, virtualization struggles with I/O heavy workloads, which are becoming increasingly important with the meteoric rise of data warehousing and distributed computing. Processors are the easiest part of the equation.
-Billco, Fnarg.com
Because you can treat a single blade chassis as a single computer, despite the fact that it has 10+ computers in it. So, rather than separate boxes tied into a SAN, you have a single "computer" with directly attached drives (SCSI drive farm) for better performance. Then you cluster piles of those with a shared SAN for what must be shared across them. Better performance than separate machines.
Oh, and blade servers have better reliability, even if you think they have more points of failure. And, depending on your setup, space is a cost that is a consideration, and compactness will save money.
Learn to love Alaska
Before reselling them they had to clean the hard drives of all the lost hopes and dreams of previous players.
It is by the juice of the coffee bean that thoughts acquire speed, the teeth acquire stains. The stains become a warning
Blizzard is a for-profit company, not a charity. Why payroll 500 of your worst developers when you can pay 50 to do the same job for a fifth the price(assuming you pay them more and benefits are a flat per-head rate)?
while(1) attack(People.Sandy);
No they are not still using blade servers, hence the reason they are selling them off as historical pieces of art basically..
"But this one goes to 11!"
vmware still has issues with all the I/O going through the hypervisor. the blades have local storage for the OS
HP absolutely promised us and our customer that there is no I/O issues using their vmWare server-solution, compared to bare-metal.....not that I believed them.
Do you understand how the concept of tax rates work? Let's say, for argument's sake, that Blizzard's corporate tax rate, is 30%. They sell a server for $250, and spend $100 of that money to ship it. They then donate the remaining $150 to charity. Assuming, then, that they could write off the $250, that means they save $75 on their taxes. So, let's look at it. They take in $250, they pay $100 for shipping, pay $150 to charities, and get to keep $75 of that. Sounds great, right? Amazing moneygrab from the company that's making about $153 million a month in subscription costs. Now, let's suppose that they were the capitalist scoundrels that you're alleging they are (given, they do a lot of things to grab money, but framing this as one of them is silly). They sell the server at auction for the same $250. They pay $100 of that to ship it. Let's assume, again, that they can write off shipping in the costs. So boom, they save $30 on taxes. $250 in income, $30 in saved taxes, minus $100 for the shipping cost. They've then gained $180. So effectively, even after the tax implications are figured, they still could have earned more money by keeping the money for themselves than giving it to charity. How dare those scumbags give the money to a hospital for kids.
from TFA:"There may never be another game as popular as WoW, and even if there is, at the very least WoW will always be considered the first mega-successful MMORPG." I'm surprised that no one has challenged this yet. I think WoW became more popular at it's high point, but I think Everquest paved the way for it, and was as popular at it is now. EQ certainly eclipsed the stuff like Ultima Online and Baldur's Gate that preceded it.
Don't moderate flamebait as Troll. Know the difference or you will be Meta-moderated.
Kudos Blizzard for doing such a decent thing, couldn't have picked a better charity.
ACs don't waste your time replying, your posts are never seen by me.
I looked at the auctions when they originally occured
they were shipped INDIVIDUALLY from France, so the price ain't that bad
Now, WHY TF they weren't shipped to one point in the US (say Blizzard headquarters) and then individually shipped to buyers-- escapes me.
every day http://en.wikipedia.org/wiki/Special:Random
It only makes sense to employ people if you have a job for them to do. If Blizzard had nothing useful for them to do, keeping them around to twiddle their thumbs doesn't make much sense.
This.
I play paper and pencil games with someone who had his department basically cut in half.
Over the past few months a lot of their tasks were made more automated and they were being sent home early due to a lack of work.
Apparently there was a cost/productivity metric that was calculated for each of them and the more expensive ones were let go.
And according to him, the severance packages were nice enough that it was clear that this was not a 'we can't afford you any more' type situation.
For example, virtualization struggles with I/O heavy workloads, which are becoming increasingly important with the meteoric rise of data warehousing and distributed computing.
And funnily enough, MMO games. When you've got tens of thousands of players buying/selling, looting, switching about items, leveling up, killing stuff, and updating statistics on just about everything they do, its no home fileserver anymore. IO was the prime issue that made the EVE Online in-game Marketplace rather unresponsive at times.
"Blades are so mainstream. People in the know use a CPU with buttloads of cores. I'd tell you what brand, but it wouldn't matter, you've never heard of it."
You sir are my hero.
+1 Disagree
I am extremely pleased that Blizzard provided a nice severance package, and extremely pleased that employees of a company that services an MMORPG are themselves fans of pen and paper roleplaying games. Cool.
I'm pro union and borderline socialist myself, but I can't see asking or forcing companies to employ people they can't use. I hope your friend and everyone else laid off finds gainful employment elsewhere.
I agree with your logic. If supply side economics worked, all of the economic growth from 2002 to 2007 would have been from something other than a real estate bubble. Not to mention the insane logic behind keeping taxes low when the nation is at war. But what do I know, I'm just some idiot on a discussion board.
Yes, but why not take those human resources and keep making product? I guess when you have all the money you will ever need for mountains of blow and acres of hookers, one doesn't care about further progress.
Take the Red Pill.
"Fun fact: As of 2010, the average number of hours spent playing WoW each week in America is 22.7"
Over 3 hours per day is average? No wonder I get my arse kicked in PvP. ;)
Am I the only one who doesn't think of pictures when someone says they are having a look at the server? I mean he literally just took pictures of the hardware?!
I'm interested in knowing which realm it was for?
Was it a US or Oceanic server?
I'm just curious.
Life takes interesting turns, but the most interest is when you're off the beaten path.
Because they're easier to manage, reduce complexity, require less infrastructure, are cheaper (once you've scaled past the break-even point, which depending on vendor can be up to an entire chassis full), have fewer points of failure, require less power and generate less heat per $METRIC and are exceptionally good for virtualisation.
Or to put it another way, pretty much everything you wrote up there is wrong. Outside of very specialist scenarios where you have the facilities in place to handle the density, you don't buy blades to save space, you buy them for all the other advantages they have over regular rackmount servers.
About the only reason to buy rackmount servers is one or more of: a) you only need a small number of them (thus making blades uneconomical), b) you have a need for lots of local disk, c) you have unusual physical IO requirements.
VMware could handle hundreds of thousands of IOPS into a single host years ago. Pretty sure it's over a million now.
Outside of exceptionally unusual corner cases, if your storage system can handle it, VMware is not going to be a bottleneck.
Mostly only in the advertising blurb.
In real life, it's nearly impossible to find anywhere that you can achieve a higher density with blades than you could with 1U rackmounts (even when you're only talking about using fractions of a rack).
There are plenty of other good reasons to buy blades, but density is rarely one of them. With that said, Blizzard are probably one of the few companies who could put in enough custom infrastructure to handle the power and cooling requirements of racks actually full of blades.
Yet blades from major vendors are cheaper than their 1U servers. Your theory doesn't carry through to real-life.
Blades are all about density. If I can squeeze 10 dual-host blades in a 7U enclosure, that's 13U saved vs 20 1U servers.
Not sure what you mean by dual host, but you can get quad socket 1U servers at a very reasonable price. I've purchased a couple. The majority of space inside the box is taken up by either RAM slots or the CPUs themselves. There's also some space for disks and fans, auxhiliary motherboard stuff (networking, I/O, etc) and a powersupply. There was really very little space to spare. I'd be very surprised if you could get significantly more density out of blades compared to quad socket 1U machines. Given the density it better, the shipping costs were lower for the 1U systems too.
Last time I checked (about 8 months ago), the quad socket machines were dense than any blades I could find in terms of FLPOS/U and very much cheaper. The installation wasn't hard, you need one power cable 1 regular network and 1 ILM network cable per U. Most of the installation time was getting the server out of the box it came in.
Incidentally, if you crunch the numbers, you will also find that a stack of 5 quad opterons is half the size, draws the same power as one of those crazy seamicro servers, has comparable computational ability, much faster per-core capacity, vastly larger single image size and a higher total memory capacity. One can't get prices for SeaMicro easily, but it would have to be very good to beat the stack of 1U machines.
SJW n. One who posts facts.
a beowulf cluster of these.
Dual hosts refer to two separate boards stashed in the same enclosure (or on the same bracket). Some OEMs call these "twin boards". Take your quad socket server, slice it down the middle and that's basically what you get.
It all depends on your application. If you need the largest single host you can get, then sure the quad-socket solution is the way to go. If you're better served by multiple smaller hosts, the dual blades tend to reduce costs and maintenance since they share one power supply stack and are designed with much simplified cabling. One area where they shine is demand-based scaling. You can have a big pile of powered-down blades drawing no power, bringing them up incrementally as needed, and this is all very easily automated. Sure, your quad-socket system can do the same with WOL, but in my experience it is much more challenging to get that working reliable, than it is to tap into a blade enclosure's built-in PDU and KVM, which are often designed with automation in mind. No need for a gazillion ILM cables and associated routing/switching complexities.
-Billco, Fnarg.com
For many companies in many situations you are right but there are lots of cases where the grandparent has a point. Look at the way Microsoft does it. They "donate" copies of Microsoft windows. This costs them practically nothing (the marginal cost is e.g. the DVD; nowadays they probably don't even have to provide that, just a sticker) but can be tax-deducted at full value. Further to that, low cost computer installations would often be done with Linux if Windows cost money, so they actually increase their software monopoly by doing it.
The key question, at this point, is what would be the cost of disposal of the blades? Most companies I have seen pay considerable amounts to get rid of old hardware. If that's true in this case it could be even worse than Microsoft. They would not just be stealing the tax-deduction from the tax payer, but also planning to dump the cost of hardware disposal on them when the purchasers finally simply throw away the blades.
This can make sense if the servers have real value and/or if Blizzard or the original vendor, HP, is going to pick up the disposal costs at the end as they should; however, all cases of corporate "generosity" have to be really really carefully looked over. This is a win even for the good/honest companies since it means they get the real credit they deserve rather than being compared with the likes of Microsoft.
=~ s,(.*),<sarcasm>$1</sarcasm>,g if any_point_you_wish();
As far as I am aware 'handling user complaints' is not the same as creating new content, and you probably want different people creating content than those who spend most of their day dealing with complaints about Chinese gold farmers.
I don't know, a game with beating down Chinese gold farmers has appeal.
Take the Red Pill.