Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
×
Databases The Internet

Cisco Introduces Rackmount Servers 93

1sockchuck writes "After shaking up the market for blade servers, Cisco Systems is launching a line of rackmount servers. But the company says its ambitions are more targeted than a full-scale 'all your racks are belong to us' assault on the volume server market. Cisco says it sees its 1U and 2U C-Series rackmount servers as offering an entry point to its Unified Computing System vision for companies who've built their data centers using rackmount servers instead of blades. But it thinks many customers will like the expanded memory capacity Cisco has built into the Xeon 5500/Nehalem EP processor."
This discussion has been archived. No new comments can be posted.

Cisco Introduces Rackmount Servers

Comments Filter:
  • Comment removed (Score:5, Insightful)

    by account_deleted ( 4530225 ) on Wednesday June 03, 2009 @09:32PM (#28205053)
    Comment removed based on user account deletion
    • Re:Sorry Cisco (Score:5, Insightful)

      by Anonymous Coward on Wednesday June 03, 2009 @09:40PM (#28205111)

      This market is already cornered by the likes of Dell, HP, and VMWare.

      One of the things that Cisco gains by doing this is the elimination of their need to re-label HP Proliant servers for their IP telephony server products, and there are many very large companies that use such, have standardized on Cisco hardware, and will buy them.

      • Re:Sorry Cisco (Score:5, Interesting)

        by swb ( 14022 ) on Wednesday June 03, 2009 @09:47PM (#28205155)

        Is Cisco actually designing the motherboards, or is it like many HP servers were, just badged boards from ServerWorks or the like?

        I would guess whatever makes it special has nothing to do with system specs but has everything to do with software loaded either into the hardware or onto the hosts that drives networking.

        • by mlts ( 1038732 ) *

          This is what I'm wondering too. What differentiates their offering from the ton of high end rackmount vendors (Dell, IBM, HP, Sun, SGI), other than the fact that it has the Cisco logo on it?

          I couldn't find much in the way of specs from Cisco's website, so couldn't tell what operating systems were offered on these systems, what management software was included, and other critical details.

          With that in mind, if I were buying a rack full of servers, I'd be leaning to a vendor who is well established in the fie

          • Re:Sorry Cisco (Score:5, Insightful)

            by Nefarious Wheel ( 628136 ) on Wednesday June 03, 2009 @11:24PM (#28205689) Journal
            Branding is important, especially when the brand name is written into contracts. There might be fewer hoops to jump through if you can single-source a solution where a single vendor is named. This lives in a place I call "procurement space".
          • by afidel ( 530433 )
            They'll have to be custom parts at the top end since the memory expander ASIC is a proprietary part they got from an acquisition.
          • by drsmithy ( 35869 )

            This is what I'm wondering too. What differentiates their offering from the ton of high end rackmount vendors (Dell, IBM, HP, Sun, SGI), other than the fact that it has the Cisco logo on it?

            Onboard 10Gb CNAs are probably the biggest differentiator. I don't think anyone else has the same memory capacity either.

            The UCS management stuff is pretty slick as well.

        • by NSIM ( 953498 )
          At least for some of the boards (the ones that support 384GB memory) Cisco has some of it's own IP on the board, but I very much doubt they manufacture the boards themselves, anymore than they do for other products.
        • Everyone sells rebadged servers. I just sold an IBM eServer 325 which is the MSI 9145 IIRC. (1U Dual Opteron/939)

      • It's not just their telephony products - in fact, those are moving to a virtualization-supported model anyway. Cisco sells all kinds of application servers that they have to buy from HP and IBM that cover all kinds of things: wireless management & security come to mind immediately. This puts them in business for themselves instead of handing money over to companies who compete with them (especially HP).

    • Re:Sorry Cisco (Score:4, Insightful)

      by Anonymous Coward on Wednesday June 03, 2009 @10:10PM (#28205313)

      You're a day late and a dollar short.

      You underestimate the power of collusion between Cisco and Gartner. My employer recently spent OODLES above and beyond that of a more reliable and featureful PBX in order to adhere to Gartner's "single source" recommendation. It doesn't matter if it costs more - it *has to be better* if it allows you to consolidate suppliers.

      It really hurts me to go in to work every day knowing that I work for idiots. Instead of a 1U PBX server that just sat there untouched for TWO YEARS, now we have 20U (6 unique boxes total) that need all sorts of poking and prodding.

      Cisco knows that they can just buy some high-end hookers for the people who make the big decisions, period. They really don't care if the quality is there. I'm there to hold the bag for them when the shit hits the fan.

      • Re: (Score:1, Funny)

        by Anonymous Coward

        Dear Cisco,

        I'm a person who makes big decisions. Please send high-end hookers.

        -The CTO

      • Have divorced themselves from reason. Gartner's purpose is to tell us what Microsoft wants us to think [slashdot.org].

      • by bertok ( 226922 )

        There's a term for it:

        Reassuringly expensive

      • by 222 ( 551054 )
        I'm going to assume that your new PBX is a Cisco solution; I am absolutely DYING to know what kind of 1U PBX you were using that required a 6 server Unified Communication Manager farm to replace? Even with Unity (Voicemail) and IPCC (CallCenter) I seriously doubt your claims.

        Something along the lines of Unified Communication Manager Express as a module for an integrated service router seems much more fitting for your needs. Who designed your current setup?
    • Re:Sorry Cisco (Score:5, Insightful)

      by mcrbids ( 148650 ) on Wednesday June 03, 2009 @10:30PM (#28205415) Journal

      You're a day late and a dollar short.

      You don't understand how this "free market economy" works, do you?

      All that's needed for Cisco is to sell more product than it costs them to make. That's called profit. So long as they make a profit, it's a good move. If they don't sell enough, then it's a short-lived moved.

      The fact that you named so many vendors (Dell, HP, VMWare) makes it clear that it's still an *open* marketplace, and that there is still competition. Thus, it's not "cornered" by any stretch. In fact, not only have I *never* purchased hardware from any of the vendors you name, one of the vendors doesn't even sell hardware! (when did VMWare get into the hardware business?)

      Personally, I welcome another hat thrown into the fray! The only possible thing that could come of this is lower prices, better quality, and more likely both. Predicting their demise as they enter the marketplace, when they are one of the most well-known and trusted brands in IT is just a tad premature.

      • Re: (Score:3, Insightful)

        by Anonymous Coward

        So long as they make a profit, it's a good move

        It's only a good move if the profit made provides the highest NPV. Or, for a simple example. If a company spends $1m on a project that nets a dollar in a year, that's a bad move, because they could have put that $1m into government bonds and gotten at least 3% ($30k).

        So profit != good. Maximum profit given available options = good.

        • Maximum profit given available options = good.

          Not even that. Any negative NPV investment should simply not be made, and if all available options are negative NPV the money should be paid out instead.

        • I was going to reply and say the same thing, but then I saw the parent. Just to expand (for the benefit of the GP):

          The concept is called opportunity cost. Basically, the if you do A, but B would have made more money, B-A= the amount of money you lost doing A = opportunity cost.

          This is, incidentally, the reason that competition in free market economies pounds out inefficiencies. If a person is efficient at programming computers but inefficient at fixing cars, then he can fix his car in less time by t
      • In fact, not only have I *never* purchased hardware from any of the vendors you name, one of the vendors doesn't even sell hardware! (when did VMWare get into the hardware business?)

        Since ESX Server.

        • Um, they sell rebranded hardware, ESX Server is software, dude.

          90% of VMware "hardware" will be Dell or HP, with some IBM thrown in. There are no boxes with "VMware" stamped on the front.
      • Your forgetting opportunity cost. If the investment they make produces less profit than the same investment would produce in some other area of their business, then it was a bad move - from a "make as much money as possible" point of view anyway.

        But I agree that CISCO isn't doing anything obviously stupid with this, they have a good chance of making a dent in the market.

    • by afidel ( 530433 )
      It's worse than that, the only server on the list that is more expandable than the servers I can buy TODAY from any of the major players is the top line 2U one. Today I can buy a DL360 G6 (1U) from HP and stuff it with 18 DIMM's and 8 SFF HDD's, in fact I have 3 waiting in my datacenter that were delivered yesterday (though they are the 4x SFF variety with optical drive, they are going to be SAN connected so I only needed 2 drives for the OS). 48 DIMM's is a technical tour de force, but I wonder if they can
      • by lukas84 ( 912874 )

        The problem is that 8GB DDR-3 memory is still far more expensive than two 4GB DDR-3 modules. And the 5500 Xeons will lower the memory speed if you install too many memory modules.

        But yeah, i can't really see why we would need that much memory in a 2 socket machine - the technology will be useful when 4 way+ configurations of the Nehalem xeons arrive.

        • by afidel ( 530433 )
          But the number of slots scale with the number of processors and the ratio of ram:cores for Nehalem works out fairly well for most peoples workloads today. Not only that but I have to question how many slots you can cram into a motherboard, a four way Nehalem EX could theoretically have 64 DIMM's without the 4x ASIC. Even if they go with a daughtercard setup like the DL585 G1 you'd still have trouble fitting it all into a 4U server.
    • by ani23 ( 899493 )
      VMWare makes servers?
    • by deviceb ( 958415 )
      With Cisco building modules that can handle normal server tasks right on the gateway... why not build your own servers. I hope they do well, i respect Cisco... the other main players I only deal with because i have to. HP, Dell... phaa
      Can Cisco enter the market with some competitive pricing is the big question..
    • I don't really understand the hostility (not really from parent, but from this whole thread).

      Cisco is a hardware manufacturer/provider/rebrander/supporter/whatever it is they do with hardware. Rackmounts are hardware. Cisco is big. What's to stop them entering into a new market? It isn't like they're trying to broach a monopoly or anything here- the market in question is relatively healthy.

      If they want to try their hands at a new product, hats off to them. They're big enough, rich enough and influential eno

    • Re: (Score:1, Informative)

      by Anonymous Coward

      You're a day late and a dollar short.

      This market is already cornered by the likes of Dell, HP, and VMWare. Feel free to try in the market place however, but I think it's a big waste of your capitol and R&D.

      You're wrong. What cisco is doing is upping the memory to CPU ratio in a very small physical footprint. This is phenominally valuable for virtualized environments or for HPC environments for any host of reasons that are obvious if you understand enterprise computing. There is not a single hardware vendor out there that has server hardware that comes close to what Cisco is now providing with the UCS platform. Show me the Dell or HP system that can have 384GB of RAM. Cisco is a big VMware partner and they are

      • by drsmithy ( 35869 )

        Cisco is a big VMware partner and they are not trying to compete with VMware, they are trying to become the preferred hardware platform for VMware environments.

        Exactly. Also highlighted by the Nexus 1000v Virtual Switch and the awesome looking "Virtual Interface Cards" for the UCS systems.

  • by symbolset ( 646467 ) on Wednesday June 03, 2009 @09:32PM (#28205055) Journal

    More RAM isn't a big deal - the 5500 series from everybody else goes up to 172GB now, and will be at least double that soon. That's plenty for now.

    The density is only 1/4th that of HP's new DL1000 [hp.com] (video).

    Interconnect is what gives these Cisco servers their shine.

    • I think you should let the market decide what "plenty" is. Many have failed predicting the upper limit of needed computer power
      • by symbolset ( 646467 ) on Wednesday June 03, 2009 @10:59PM (#28205575) Journal

        I'm not Bill Gates. 640K might have been enough for anybody back then, but if he had only said "for now", we wouldn't be having this talk. I have opinions and I'll share them. Most of the time after a few years the market agrees with me.

        We're in the technology singularity. Stuff has already gotten silly and it's about to get absurd.

        Long before the aforementioned RAM quantity becomes a bottleneck for 99.9% of uses you're going to need faster RAM, a faster CPU (or more CPUs) to talk to it, more channels to talk to it with. We're half a year away from 8 cores per CPU, and 9 months away from 12 cores at most and those platforms are going to come with more RAM channels, and hence even more RAM per server, even without considering that DIMMs are going to hit 16GB soon. Likely it will be much sooner. Between now and then we'll need faster interconnects for inter-node communications, faster storage like this [engadget.com], and faster networking like FCoE (tomorrow, literally). As much as I hate the waste of throwing out year old servers, software makers are making it an imperative by insisting on licensing that defeats the technology value proposition. It may not even be wasteful as each server increment does twice as much with half the power. People who use this stuff are well paid to replace the hardware that lives under these limits frequently because the software costs at least 4 times as much as the hardware.

        /and yes, if you use open software you don't have this problem - but you're usually paying per server for support, and that amplifies the incentive to throw out your old gear every year.

        The economic contraction has turned out to be the harsh winter that brings forth a summer of great fruit. Everybody in the trade is emptying their cupboard of innovation in the hope of gaining market share, rather than holding it in reserve for a rainy day. Because it's raining now.

        What we need now is services that need this extra gear. If somebody doesn't come up with it soon Google's going to shrink down to 90 individual racks in somebody else's datacenters - three per geographic area.

        //And no, we're not dumb enough to burn these cycles running the server version of Vista. We get paid to be useful.

        • The economic contraction has turned out to be the harsh winter that brings forth a summer of great fruit. Everybody in the trade is emptying their cupboard of innovation in the hope of gaining market share, rather than holding it in reserve for a rainy day. Because it's raining now.

          So if Cisco is Richard III, Bill Gates is King Edward IV, and Lady Anne is a stand-in for the purchasing managers of corporate America, who's playing the role of Henry Stafford?

          Seriously, somebody mod this guy up. Bonus points f

        • What we need now is services that need this extra gear.

          Welcome to life after the financial starvation of the NSF; and corporations dumping all internal R&D with a financial return in the long term. The Current White House Administration is throwing money out the window to address this issue; but lacks the specific focus necessary to properly address the specific problems.

    • "172GB ought to be enough for anybody!"
    • by Euzechius ( 600736 ) on Thursday June 04, 2009 @04:10AM (#28206895)

      I work for Cisco, so this post is biased.

      If you want to know more about Intel Nehalem 55xx architecture [zdnetasia.com].

      It explains that a the server manufacturer using the Intel Nehalem 55xx processor can support up to 3, 6 or 9 DIMMs/socket. This corresponds with a memory bus speed of 1333, 1066 or 800Mhz. The latter is not often implemented and would give you (9x2x8GB) 144GB in a dual socket system.

      What Cisco did is, developing a patented "memory switch" which presents up to 4 DIMMs as 1 to the processor, MULTIPLYING THE ALLOWED RAM TIMES FOUR. If the memory is running at 1066Mhz this gives you 48DIMMs. If the memory is running at 800Mhz this would allow up to 72 DIMMs in one server. The latter one has not been implemented.

      Where would you ever need this kind of memory?

      * Running VMware ESX, XenServer,... and assuming 3-4GB per VM -> imagine 96 VMs per physical box
      * imagine running a 300GB MySQL database out of RAM without the need of a high end machine

      Also the price per GB is not linear for memory. 8GB costs currently way more than 4x 2GB. So if you still don't need the 384GB memory, you can fill the 48DIMMs with 2GB and have a 96GB RAM server for a lower price.

      There are also a lot of other features which are really different and better than the competition, such as centralized management per 320 servers. In more enterprise environments customers can also consolidate their SAN and their LAN network by using open standard FCoE.

      Please check it out at Cisco - Unified Computing System [cisco.com]

      • by TheSunborn ( 68004 ) <mtilsted@NoSPAm.gmail.com> on Thursday June 04, 2009 @06:01AM (#28207329)

        How much (If any?) extra latency does the switch add? It's not the first time someone tries something like this but the latency normally might be so bad that you might just want to buy an other server instead. (Unless it's a database server, because even slow ram is much faster then the disk :}

      • ...consolidate their SAN and their LAN network by using open standard FCoE.

        Oh joy, another new connection method with all new connection hardware with all new expenses and all new training. I still haven't reached effective ROI on the FC installation. Please tell me why I need to incur this expense?

        • by afidel ( 530433 )
          Because in a large datacenter you can save millions on reduced wiring, cooling, devices and ports. It's really only for greenfield installs but you'd be very foolish NOT to look at it for a new rollout.
    • I don't get their maths: Cisco has boosted the memory capacity of the Xeon 5500, using a custom ASIC (Application Specific Integrated Circuit) to provide the processor with a four-fold increase in the number of memory modules it can access. This expands a UCS Xeon 5500 system from 144GB to 384GB, and also gives users the option of using more affordable memory configurations.
      144 x 4 != 384

      For most people the practical limit on memory is simply the price of the higher capacity DDR3 dimms - 4GB ddr3 stick
      • 4GB ddr3 sticks are only just becoming affordable and 8G ddr3 sticks are hugely expensive!

        And 72GB for 8 cores is a good bit for now. In a year the new platforms will have more slots, and the big stick will be 16GB, so no worries.

  • OEM HP servers? (Score:3, Insightful)

    by c0d3r ( 156687 ) on Wednesday June 03, 2009 @09:47PM (#28205159) Homepage Journal

    Last I knew, Cisco was selling products using OEM'ed HP servers. Sure they aren't just HP servers?

  • Take that, HP! (Score:5, Insightful)

    by binaryspiral ( 784263 ) on Wednesday June 03, 2009 @09:51PM (#28205185)

    HP used to provide hardware for Cisco's appliances and servers that they resold as Cisco branded gear... Call Managers and the like.

    Well, HP's been really pissing off Cisco by selling ProCurve switches with lifetime warranties and converting Cisco Catalyst switch users over to HP ProCurve customers. Cisco's been losing all this SmartNet gravy that they wallow in year after year. So this is their answer... sell servers to piss in HP's very large bowl of Cheerios.

    Good luck Cisco, you're entering a cut throat market with well established hardware vendors in a global recession... You've either got a large pair of brass balls or you're just really really stupid.

    • Re:Take that, HP! (Score:5, Interesting)

      by BBTaeKwonDo ( 1540945 ) on Wednesday June 03, 2009 @10:04PM (#28205265)
      When a company has over 30 billion dollars in liquid assets (Excel warning) [cisco.com], entering a market that's closely related to the one it's currently in does not classify as ballsy, even if said market has competitors.
    • Re:Take that, HP! (Score:5, Informative)

      by teh_c0unt ( 1392683 ) on Wednesday June 03, 2009 @10:10PM (#28205311)

      Good luck Cisco, you're entering a cut throat market with well established hardware vendors in a global recession... You've either got a large pair of brass balls or you're just really really stupid.

      is cisco not a well established hardware vendor? http://www.thestreet.com/story/10508379/1/tech-rumor-of-the-day-juniper-cisco.html [thestreet.com]

    • I hear that it's game on! Cisco intends nothing less than openn war with all its server vendor partners including HP, IBM and Dell.

      So that's an easy short. Who wants to bet against HP, IBM and Dell? To bet one against the other is arbitrage. To bet against all of them at once is just dumb.

    • by fwr ( 69372 )
      Certain Cisco switches have limited lifetime warraties now also.
    • by comm3c ( 670264 )
      Heh... this isn't just about the network gear, this is also about HP's acquisition of EDS.

      You see, once upon a time, HP left their network gear alone. Cisco and HP weren't much of competitors. Then, one day, after firing a CEO, HP decided that they should try to push the product line.

      In the mean time, Cisco wasn't worried. But Cisco also had this "strategic alliance" with EDS, complete with multiple teams related to the account. One day, EDS got sold to HP. HP is pushing its product on EDS customers.

      Cisco

    • Cisco's been losing all this SmartNet gravy that they wallow in year after year.

      Because the person who knows how to use a spreadsheet has figured out SmartNet is a financially dumb move. Networking technology advancement is accelerating; but the SmartNet contract will only provide an equal replacement. For example: why spend the equivalent money to get replacement 10/100 equipment when a properly planned replacement scheme can upgrade your network to 1Gb for the same cost?

      HP is not the only network gear company taking market share from Cisco. Basic LAN networking gear from the Other G

      • by dawich ( 945673 )
        SmartNet is only dumb if you pay too much for it. TAC is still way too useful to drop SmartNet and self-insure.
  • by TD-Linux ( 1295697 ) on Wednesday June 03, 2009 @10:19PM (#28205355)

    Seriously, Cisco? Yet another boring Xenon server? There are so many out there I can't tell the difference.

    You could have done something unique and interesting... throw a couple ARM Cortexes into a ultra-low-power 1U server... and make it completely redundant, just for kicks. Or you could have integrated something you are good at, like, well... I guess that option is becoming slimmer.

    Anyway, cheers for yet another undistinguished product entering a crowded market aimed at legacy users with falling demand.

  • by sethstorm ( 512897 ) *

    It'll require an expensive support contract just to load any software on it or add any new hardware to it.

    • Re: (Score:2, Insightful)

      by cfryback ( 870729 )

      It'll require an expensive support contract just to load any software on it or add any new hardware to it.

      Based on what we have paid for a very basic ASA 5510, I shudder to think what the support costs are! And to paraphrase other people, we have a HP cClass Blade system, that has proven to be very reliable and econmical. Did I mention that we are an HP shop? Replaced all of our 3Com switches with ProCurves, for 1/5th the cost - all with lifetime warranty and software/firmware updates as needed - all for NO ANNUAL FEE!

    • Not quite; if you load any software on it not installed by the vendor (say, security patches) then you void the support contract. At least, that's how they do it with their blue HP servers.

    • Just like with all other servers in the business world...
  • Cisco.... (Score:1, Redundant)

    by snowblind ( 175857 )

    ...jumped the shark when the changed to the "cute" logo. Now this is just them introducing a smart mouthed kid to make us chuckle until they end up in re-run hell.

    What do you do when you're at the top of your market share and can't innovate to come up with something else and you've bought up every home market vendor?

    Crash in on a new commodity market with an overpriced product; while maybe considered a caddilac, sell it at Mercedes prices.

    All the while Juniper's chomping at your heals in your traditional s

  • by Bluecobra ( 906623 ) on Wednesday June 03, 2009 @11:27PM (#28205701)

    I think this is a great thing for Cisco. Okay, so nobody will buy their servers for regular stuff. But they will buy Call Manager servers and the like. At work we have 3 Cisco servers that are re-branded IBM boxes. One is for our Unity voicemail system and the other two are for Callmanager. When there are hardware issues, I need to call Cisco who then calls IBM to fix it. I think from a support perspective, it would be a huge benefit to actually MAKE the servers you are supporting that way support requests get processed more efficiently. Cisco doesn't just have IBM servers either, they have HP as well so that would be two vendors that they don't need to deal with anymore for support.

    • by afidel ( 530433 ) on Thursday June 04, 2009 @12:10AM (#28205895)
      Yeah instead of working with HP/IBM to support their servers they are going to have to train all their field engineers how to properly service and diagnose their new server line. Oh and IBM announced a big deal with Brocade for FC gear the day after the launch of Cisco's blade servers, I wonder how much revenue they will make on the servers vs what they are going to lose in other markets by pissing off all their partners. Unless you have a lot of clueless CTO type get snowballed by the Cisco rep I really don't see these things going anywhere.
      • Well IBM is no longer a "partner", but rather a partnering-competitor type. HP has always been a double faced company, since they are direct competitors of Cisco, the pro curve line.
  • Is that the 40 core version of Xeon 5500?

  • by speculatrix ( 678524 ) on Thursday June 04, 2009 @04:16AM (#28206923)
    I'm surprised Cisco didn't simply buy Sun Microsystems - a reputation for making expensive, over-engineered hardware (both).

    It's only a small step for Linksys to move from making NASs and media players/extenders to PCs, so I expect we'll see a Linksys version of some of the small eee desktop etc.
  • These new 1U/2U servers may appear to be even less interesting than the California (UCS) products that were introduced earlier this year. However, the reality is that Cisco is actually rounding out the lower end of its server product line before they introduce the much more innovative, higher end (higher margin) SMP systems later this year. Intel just announced the new SMP platforms based on 8-core processors that scale to 4 and 8 sockets. Expect Cisco to scale the socket count even further. Going into the

Never test for an error condition you don't know how to handle. -- Steinbach

Working...