Please create an account to participate in the Slashdot moderation system

 


Forgot your password?
Close
typodupeerror
×
Hardware

Linux Based Router 90

Posted by CmdrTaco from the only-a-matter-of-time dept.
Troy Larsen sent us a link to a yahoo press release proclaiming a company selling a Linux Based Router. Up to 40 gigs per second and up to 128 100mb ports. Yum.
This discussion has been archived. No new comments can be posted.

Linux Based Router More

Linux Based Router

Comments Filter:

  • Not that new (Score:1)

    by Anonymous Coward
    Open software has been used in gigabit-class
    routers for quite some time. One of the
    very first -- BBN's gigabit router -- used
    BSD4.4.

    It is important to understand that the OS
    doesn't do the packet forwarding. That is
    done by specialised ASICs. So the speed of,
    or minor flaws in, the OS's IP stack are not
    of much relevance.

    What is more important is that the OS can
    keep the routing processes running. These
    processes exchange the route information
    and load the hardware forwarding table that
    the forwarding ASICs then look up.

    There's a lot to be said for the reliability
    of proven OSs and proven routing code rather
    than building from scratch.

    One of the major out-of-the-Labs applications
    of UNIX in the early years was to supervise
    the operation of telephone switches, so we
    seem to be coming full circle.

  • One hell of a hacked IP stack (Score:1)

    by Anonymous Coward
    do you have any useful comments to make about my post, or are you just going to criticize my ficticious example?

    Let me clarify what I was trying to say:

    1) NFR depends upon being able to read() a buffered stream of packets in order to do effective packet capture

    2) Linux doesn't implement this

    3) Therefore, Linux will perform very badly running NFR

    4) The immediate conclusion to be drawn is NOT that Linux's TCP/IP stack sucks. (in fact, packet filtering has NOTHING to do with TCP/IP or any other network protocol)

    5) Likewise, the fact that there is no DECnet support in OpenBSD, but there is in Linux 2.2, does NOT mean that OpenBSD's networking sucks.



    Do you have anything useful to add?

  • Open source router - QNX (Score:1)

    by Anonymous Coward
    QNX/Neutrino always struck me as the way the HURD should have been ( except, obviously, QNX is not Free Software)

    QNX is incredibly robust. Maybe we could gravitate linux to protected device drivers, at least on intel and PPC - say ring 3 user space, ring 2/1 device drivers and modules , ring 0 kernel ?
    ( I may have the ring ordering back -to -front)

  • One hell of a hacked IP stack (Score:2)

    by Anonymous Coward
    > That and some recent NFR results showed that
    > linux dropped 90% of the packets on a 47mp line
    > whereas the BSD's dropped 5% and Solaris on an
    > Ultra1 dropped > 10%.


    NFR is written specifically for the Berkeley Packet Filter using read(), which Linux doesn't implement (Linux supports filtering but doesn't buffer the incoming packets to allow read()). End of story. Why don't you compare the speed of Linux 2.2 DECnet versus the DECnet in the latest OpenBSD while you're at it?

    However, if you look at the latest patches from Alexey Kuznetsov on:


    ftp://ftp.inr.ac.ru/ip-routing/lbl-tools

    you will discover a kernel patch and a patch to libpcap labeled "turbo". This patch implements packet capture via a ring buffer shared between kernel and user space. This is something even BSD doesn't have- they must go through an additional buffer copy in the read() call.

    This patch also eliminates 2 out of the 3 main problems with the Linux packet capture code outlined in this message from the NFR mailing list:

    http://www.nfr.net/nfr/mail-archive/nfr-users/19 99/Feb/0110.html

    The remaining problem, #3, may or may not still be there, but only involves copying a few bytes here and there and so isn't a very big deal if it does exist.


    It will be interesting how the different OSes stack up doing packet capture in the future.
  • That'd be these guys [linuxrouter.org], the whole shebang fits on a floppy.
    It runs off a ramdisk and has a menu based front end.
    One reason Debian rocks is because the distro is broken down by licensing.
    The main distro is strictly GPL software. It's also the most thoroughly tested.
    Personally, I like SuSE on my desktop,
    but IMNSHO Debian is _the_ way to go on a server.
    Now RedHat, OTOH...thats the AOL of Linux distros,
    a good place to start, but after a while it's time for greener pastures.

  • Debian Based? (Score:3)

    by Aaron M. Renn ( 539 ) <arenn@urbanophile.com> on Tuesday April 06, 1999 @12:46PM (#1946799) Homepage
    I seem to recall that someone was building a Linux "router distro" based on Debian. Is that this company? I also hear that Debian thinks it is good for people to be taking their main distro and tweaking it for various special purpose apps. Are there any others I'm not aware of in the pipeline?
  • Posted by The hacker:

    I think it is extremely great to have a Linux based router building by a big networking company.
    I am sure they have BGP in their product since I used their (Nbase-Xyplex) fiber optic & Switches. They have all the features you just dream about in their product and it works just astonishing. Thus I am sure they will have BGP-4, POS and WDM in their Linux-based router.
    I think that Linux-based router with a lot of features, an high performance and so scalable is a innovation to our ISPs community. So I think the idea to keep eyes on this product and watch it is right on place.
  • Posted by The hacker:

    They do it again, after the first 10/100 ethernet switch they come again with the first Linux-based switch router - WOH.
    I think it is extremely great to have a Linux-based router building by a networking company.
    I am sure they (Nbase-Xyplex) have all the features in their product since I used their (Nbase-Xyplex) Fiber optic & Switches and they have all the features you just dream about in their products and they are always preceding the market.
    I think that Linux-based router with a so many features, a high performance and so scalable switch router is a innovation to our ISPs community. So I think it is a good idea to keep eyes and to say thanks to Nbase-Xyplex for given us the ability to role our world with an Open router.
  • I know a Cisco Pix (high-end firewall) uses a P2-233. Also, it uses 32 or 64 MB of RAM and uses flashram cards for storage. The OS is probably living in a flashram too.

    I wonder how much non-pc hardware is in a Pix. On the other hand, it can take a maximum of 4 interfaces with a 155 mbit ATM interface being the fastest it can handle. A properly configured Linux box could do the same job.

    I haven't seen the GUI tool for configurating the Pix, but the text-interface (via telnet) sux and is more complex than ipchains. Also, the documentation leaves something to be desired. I think I would prefer a Linux box with some reliable hardware.

    The Pix does have fail-over capabilities. Something similar can probably done under Linux, but right now I haven't a clue. Anyone any ideas?

    Mathijs

  • Which explains why FreeBeasties only used to quote loopback numbers to compare to the new Linux stack. "Ooh! FreeBSD is faster over loopback than Linux over the wire. Film at 11." -- Linus T.

    They're both good, and my P100 with an EEPro 100 card has no problems saturating a 100BT line.

  • 1x is around 150kb/s, 4x should be around 600kb/s plus NFS overhead... it'd put your network a bit over halfway.

    I've never personally ran into a situation where network performance from linux has suffered, and I use boxes which push TBs/mo in web traffic, millions of hits a day with CGI and no complaints (the boxes themselves usually don't even bust a load average of 1, unless some major disk access is slowing things down). I'd expect that people who're selling routers with linux in the OS wouldn't get very far if there were bottlenecks as bad as some people claim.

    --
    rickf@transpect.SPAM-B-GONE.net (remove the SPAM-B-GONE bit)

  • Zero copy transmits are possible with certain NICs and drivers.
  • Nice FUD moron. If you're jealous of FreeBSD, don't whine about it.

    That wasn't FUD. He said "both are good" (after quoting a sarcastic Linus comment, granted).

    As for the comparison, I share his experience. One of my boxes is an ancient 90 Mhz Pentium running Linux 2.0.x, and it has no problems at all saturating my 100mbps ethernet. E.g. I use it to burn CD's from a master over the net at 4x, which I believe is pushing the limits.

    I've kept the network and that box otherwise fairly quiescent while doing so, figuring there's no point in asking for coasters, but still, it's handling the net, the scsi device, and the CD burner software all at once.

    I personally have no idea if FreeBSD/NetBSD/OpenBSD are better or worse in these regards, and it doesn't matter to me (except for being interested in knowing when to recommend one thing or another for various purposes). If Linux didn't exist, I'd be using BSD; I have a personal interest in it.

  • I've been using, installing, and recommending Xyplex equipment for years. It is without exception well-designed, intuitive, and based only on open standards. They're a good company with good products - it's a shame that they have to live in Cisco's shadow. I'm excited about this new product and hope to get a chance to tinker with one soon. There are just so many possibilities this opens up...
  • Nice FUD moron. If you're jealous of FreeBSD, don't whine about it.
  • Who gives a damn about DECnet? Most people don't. Show me a major internetwork routing DECnet over itself and then you'll have a reason to yack about nonsense like that.
  • "Amazing what you can cram into 1.44MB"

    Not _100%_ sure about the LRP, but most mini-distros I have played with actually cram it all into ~1.7MB on the floppy that's been formatted for a very high density.

  • I've always been perplexed by people saying "xxx distribution is ok to start with, but no one else should bother with it." (Usually directed at RedHat.)

    Let me freely admit that my main Linux machine was originally a RedHat box. Why? I was new to installing linux, for starters. But I also wanted something I could loan to interested friends and relatives that they might be able to install on their own.

    I have never bought another distribution. I may pick up Debian or Slack some time in the future, just to compare. But here's my point: My box is Redhat originally, but I feel I have become an intelligent user (compiling the kernel and other software, learning how to manually setup services, boot configurations, etc.). Why would I now go out and buy any other distribution as a "graduation" from RedHat? My machine is my customized version, I have already gone beyond any distribution in terms of what works for me.

    (I can see for specialized uses like LRP it is quite a good idea to use another distribution.)

    +LO
  • I have been running a linux router ever cince the net package has been out for Linux!!!!!!
    THWAP! THWAP THWAP!!
    ok people, getting all wet over something that is just a jumble of hardware with a great OS gluing it together isnt something good, it's olllllld news and just a marketing hype.

    It's cool that linux is in the mainstream, it sucks as this company get's credit for what most of us have been doing in our sleep for years.
  • a router requires stability and near perfect up-time... something NT hasn't been able to do, cant do currently, and is not expected to be able to do for years to come.

    if you have mission critical, you dont use NT.
  • "Linux is only free if your time has no value" - Jamie Zawinski

    Great, that means I paid twice for all this NT that eats up all my time. Uhhh...

  • If you're worried about the backbone, you aren't running cisco - you're running MP Wellfleet kit so you can sleep at night.

  • being employed by a xyplex-nbase reseller, i read
    the article with mixed emotions. yes, it's nice to
    see people using linux and oss to build these kinds of products, but you must consider *who* is
    building the products. having worked with the
    xyplex gear before the mrv buyout and subsequent
    merge of the xyplex and nbase product lines, i have a place in my heart for the xyplex chassis solutions (network 9000's). yes, they're slow and
    limited, but i sleep well at night knowing that we have 100 or so of the 15-slot chassis out there at
    customer sites. historically, the xyplex products have been rock-solid. but, with the merger of the two companies, the new products leave little to be excited about. anything that comes from nbase is inherently flawed, from my experience. these people can't seem to make a product that can gracefully deal with spanning tree. how can they be expected to create something that's competitive with today's gigabit switch-router manufacturers. unfortunately, most all of the new xyplex-nbase product line is developed by nbase engineers. so, being the devil's advocate that i am, i'll just have to wait and be amazed if they actually make a worthwhile product.

    ..ok, now i'll step down from the soapbox.....
  • I dunno about the others, but on my OpenBSD box, also a P90, ftping a file from it cause X to simply STOP until the file is done. A simple ping flood will cause the mouse cursor to jerk around and make it hard to move windows around. Maybe some BSD advocate will tell me what I'm doing wrong, but this is not the searing network performance that I was told to expect from the BSDs.
  • I expect that this would be a target market for the embedded version of NT that MS is working to bring out.
  • We will see what this really is, but it is already clear that this is not the equivalent of a 486 with a couple of ethernet cards running IP forwarding code.

    Dig it?
  • How about a couple orders of magnitude performance difference with the LRP as the looser.

    Don't get me wrong, LRP is absolutely great!

    It meets the needs of millions of people, businesses and organizations who need a reasonably flexible packet handling device running at T1 to ethernet speeds but there are plenty of applications for purpose-built hardware solutions.
  • A reasonably priced switching fabric under the control of open source code which and have modules added to it could be very cool.

    Existing Linux based routers are great for the low end, but they can only scale to a certain point because the CPU is involved in all packet forwarding.

    In something like this, I/O porcessors, or some specialized ASICs do the scut work of moving data around making low level routing decisions. The OS and CPU only deal with managing the switching fabric. They generally only get involved when things change, such as when a packet comes in for a new destination, or when a route to a destination changes, etc.

    Having this management layer available for manipuation could enable linux developers to bring out a new class or network applications.

    For example, if this hardware is cheap enough and presents clear interfaces, it should be relatively easy to implement a high performance URL aware HTTP load balancer which could eclipse the performance of any sort of software/general-purpose hardware solution by multiple orders of magnitude.
  • I was going to say that this would be pretty expensive, but given that this router lists at $300,000, the I2O route could be quite practical.

    The I/O processor would forward packets based on a local flow cache. Packets not matching the flow cache would generate an event which would be handled by a module running on the linux OS.

    Incoming HTTP requests could be forwarded to a linux module which would spoof the handshake to get URL information. It would then open a connection to the apppropriate web server, populate the flow tables with the appropriate entries and then hand the flow back to the IOPs. The IOPs would forward packets and rewrite headers as needed.
  • the short answer is no. I don't think that linux would like trying to manage the cbus, pos cards, vip-2s, route switch processors, etc.
  • Maybe we can use it to convince Cisco to hand out the source code for their 12.X IOS. That is some code I would like to get my hands on.

    Unfortunately, the product doesn't seem to be up to snuff yet. No talk of OC-48 Packet over Sonet, or even OC-12 POS, no talk of GigE, no mention of BGP??? Looks like one to keep your eyes on and watch how it develops ... exciting, but not ready for primetime.
  • You were led to believe correctly. The deeper question that yours leads to is whether or not any operating system's IP stack is up to speed with a custom-designed solution. Juniper, I know, is making some gigabit-class routers based on the BSD kernel, but nobody has been able to take away Cisco's IOS market share yet ... is it worth our time to try to route with OS-based systems? Is a hacked version of routed or gated robust enough to pass 20 or 30 gig across a switching and routing backplane? More power to them if they can, but i have doubts.
  • pardon me replying so much, but I work on Ciscos all day and am interested in any other players in the market. A couple of the issues I see are:

    -- express/optimum switching. Here, we are using the first packet headed toward a destination over a certain port to evaluate through the router's access lists, and then allowing the remaining packets toward that destination over that port to flow through the switching hardware without being evaluated by the processor? Is a BSD or Linux-based gated able to handle the logical concept of flows to optimize access lists and route processing?

    -- nifty features: yes, they're standards based, but does the gated gsr support soft inbound and outbound soft reconfigs on bgp? What about nifty things like HDLC, which beats the hell out of PPP. ISL trunking between your switches and routers? Fast Etherchannel? Fast drops on access lists, which saves immense amounts of processor time when you are writing a smurf filter. Easy disabling of directed-broadcast, while we're talking smurfs. Rate limits on circuits?

    What I'm saying here is that a BSD or Linux based router is going to have to sell itself to a lot of people who maintain internet backbones for a living. It is going to have to have a feature set that meets or exceeds our current vendor's. While I can see the applicibility of a BSD or Linux based router on the low end, I have yet to see an entry that I would trust a nationwide backbone to.
  • A little more than processing OSPF maps is taken care of in software:

    -Processing of access lists

    -Managing switching path logical connections

    -Maintaining all processing for your Interior Gateway protocol(s) of choice

    -Maintaining all BGP route maps, route tables from upstream neighbors, route reflection client or server processes, metric processing, etc.

    -Maintaining the master route table, where BGP and IGP routes are held and routing decisions are made

    -Any and all network management

    -Any and all console diagnostics, line card monitoring, environment monitoring, power monitoring, etc.

    -Handles any high-level encapsulation, especially tunnels and encapsulation of Appletalk, DECnet, IPX, X.25, etc.

    That strikes me as a little more than a configuration front end.
  • I believe the Linux Router Project ( http://www.linuxrouter.org/ [linuxrouter.org]) may be what folks are looking for here.... the idea is a mini-Debian distro-on-a-floppy. 'Tis pretty slick, what little I've played with it. Amazing what you can cram into 1.44MB. (Reminds me of the old days of the 255-byte BASIC games :) :)

    der rezident old fsck...
  • I have been led to believe that BSD was far superior to Linux when it came to networking performance. Why base it on Linux when BSD is still much faster?
  • Actually, the OS on a gigabit switching router (GSR) doesn't really do much. No OS is going to actually keep up with 40Gbit/sec on modern processors, the real work is done by the hardware with the OS providing GUI functions and poking the appropriate hardware bits as needed. I'd guess that using Linux is going to do them more good from a marketing standpoint (riding on the current wave of hype) then from a technical one.

    A similar product, the Juniper Networks M40 [juniper.net], uses a modified version of the FreeBSD kernel.

  • Hell 'we' get 400 Gbps on a single fibre.

    But then we're the reason we say *nix!

    Drool [bell-labs.com]

  • Here's a question for anyone who's worked with both Linux and IOS:

    Would it be possible to compile Linux to run on the Cisco equipment? IOS supports downloading a new "Flash ROM" for IOS updates, so I'm guessing that Linux could run on a Cisco router/switch/hub for a lot less than the several-thousand-dollar IOS software?????

    Please enlighten me...
  • Cisco licensed QNX last May, so I don't think they're going to use Linux or go Open Source. Check out QNX's press release [qnx.com].
  • You may also want to check out another mini-dist [fireplug.net] based on Debian that is aimed at a firewall/routing box booting from one floppy.
  • Currently, both are faster than typical hardware. Neither is fast enough to route all that on a 1999 processor.

    Linux can beat BSD much of the time. See the scoreboard [rutgers.edu], where Linux toasts both SunOS 5 (SysV-based Solaris 2) and SunOS 4 (BSD-based Solaris 1) on Sun's own hardware.

    I'm sure the OS is just there to config the hardware and perform low-speed tasks as needed. The real work is done on special-purpose chips. Just think about the memory bandwidth your CPU would need to handle all that.

  • Yeah, that Linus guy must be a dumb newbie.
    Only a dumb newbie would be seduced by such
    a simple no-hassle distribution like Red Hat.
    Linus probably has Linux just to be cool, and
    really only uses Windows 95. Yeah, Red Hat users
    like Linus might even have trouble with AOL.
  • >``The OSR8000 has addressed the market demand for a flexible, high
    >performance enterprise solution that increases network bandwidth on demand
    >and preserves mission critical application requirements for enterprise and
    >service provider networks,'' said Noam Lotan, president and CEO of MRV
    >Communications.
    >

    Do you think he actually *said* that? Wow. It seems like it would take a commitee to say something like that.
  • This is the same company that stole graphics from my employer's web site for there AcceleRouter product. Granted, my employer wanted to look into switching to Linux from another embedded OS, unfortunately Linux is just too resource intensive for our box. As far as stacks go, BSD and Linux stacks have their strengths. One thing I would like to see in the Linux TCP/IP stack is the ability to perform zero-copy transmits and support hardware TCP/UDP checksum calculation. What also would be useful would be the ability to blast data out of the disk cache without copying. Unfortunately it will take some work to adjust Linux's stack architecture and all of the network drivers.

    Linux has a nice TCP/IP stack, but it's not as feature-rich as BSDs. I.E. no T/TCP or RSVP (last I checked).
  • What makes you think that Cisco IOS has an advantage in performance over Linux? IOS can run on a single processor, Linux can run on multiple processors.
    I'm not much of OS expert though...Any gurus out there:-?

  • There is a neat project going on named Linux Router Project [psychosis.com]. It was described in Linux journal [linuxjournal.com] I think March issue. The cool thing is that the kernel can fit on a 1.44M floppy. Definitely NT can't do that.

  • Well one of the obvious is that LRP is free, Open Source and the other is for sale and one can't get the source. Another is that LRP is not aimed at businesses, more at hackers. (yes hackers as I and other ./ers define it, not the way media does).

  • From: Bob von Ende
    NBASE-XYPLEX
    Bobv@nbase.com

    In answer to many questions and comments I offer you some points of clarification concerning our New High End Linux Router.

    Architecture - Component based

    Open Interface
    Flexible, high performance switch fabric
    Back-end Linux engine

    Scalable Architecture
    All components are open; each component can be updated, replaced, or changed independently.
    Any of the components are open by themselves.
    Any of the components can be upgraded independently
    User programmable ASIC's are open and extensible

    Open Interface

    Open, flexible & extensible interface
    Open application programmer interface provides for direct user enhancements and upgrades.
    A non-proprietary Operating system which eliminates the single vendor dependence and encourages in house development capabilities.
    Users can add their own features or obtain additional features from the developer community.
    New features will become available quickly and be certified and supplied by the vendors faster than other competing router OS's.

    The Switching Fabric

    An Active backplane with (Qos)
    Scalable 40 Gbps to 160 Gbps cell based switching fabric providing 26 million pps throughput.
    Multiple processors with 4 Intel processors upgradable to 16 processors
    Native multicast support
    Frame & Cell Switching
    Classification upon any frame/cell pattern

    The Back-end Linux Engine

    Functionality: routing protocols & any other network applications
    Per user customization
    Unlimited & fast adding/updating features and applications
    Easy hardware expansion & modification



    Conclusion

    Non-proprietary router OS
    Seamless Integration of custom applications
    Flexible, scalable, high performance solution
    Full security functionality
    Features and capabilities added faster than competing products
    Support for most routing protocols
  • We had the linux router working great, And then we had the linux brige working also. Most stuff i have found works awsome with linux, If i could just get bewolfe to werk right. Yadyad, Check out vhosts.net [vhosts.net], Webhosting for only $60.00 per year. Unlimited hits, Just email admin@anvdesign.net :)

Slashdot Top Deals

As long as we're going to reinvent the wheel again, we might as well try making it round this time. - Mike Dennison

Close