Slashdot Mirror

Can you point me to a link? According to this, the VPN 5001 can be swapped for a 3030. The itemized list of clients on the 3000 series page doesn't include Mac.

Part of what makes RF radio stations economical--and even occasionally profitable--is that the marginal cost of providing the broadcast service to an additional listener is essentially nil, modulo geographic saturation and transmitter power.

Today's streaming media services, however, incur a high marginal cost per additional listener--cost scales linearly with the number of listeners. There have been several attempts (Akamai, RBN) to get listeners to use a "nearby" transmitter, but these only flatten the cost-per-additional-listener line a bit by saving money close to the originating transmitter.

The Internet evolved a more bandwidth- and cost-efficient distribution model years ago in the form of multicast, but it was never widely implemented in enough of the places where it would have made a difference--backbones, routers, terminal servers, DSLAMs, cable companies, etc.

The idea is that a multicast packet stream should have a very small bandwidth footprint for the most expensive parts of the trip from transmitter to the receivers, only needing to be duplicated at the last few legs of the trip, where receivers aren't on the same physical network.

IOW, no matter how many of an ISP's customers are listening to a multicast stream, the ISP only has to transfer the packets from the expensive Internet once, and then make sure they get routed down the cheaper links to those customers who are listening.

Now that NAT is becoming more and more widespread, the situation doesn't look good--but hopefully IPv6 will kill NAT, and improve the multicast situation by opening up a vastly larger range of multicast addresses, and therefore a larger maximum number of simultaneous multicast connections.

Some fun links:

An Introduction to IP Multicast Routing (from Google cache, the site seems to be down)

Some stuff from Cisco

RFC2375: IPv6 Multicast Address Assignments

IPv6 Multicast Standards

Of course, everyone who adopts Seibel's software does this.

Wow, I gotta get me some of that!

Heh, except RHCE and CCIE. I've seen *competent* people walk into an RHCE exam and come out wondering why they're even alive...

http://www.cisco.com/warp/public/116/t1_flchrt_mai n.html

Flowcharts, etc. T1s are hard if you make them hard. Or if you use Bay routers :)

Put flame jacket on... Let's face facts, people. The fairest way for these moviemakers and musicians to get their royalties IS through levies on blank CD-R, CD-RW and DVD-R. I know for a fact that when the majority of people go out and buy a CD recorder, they're thinking "I need a place to put my pron, warez, music and video-CDs" NOT "I need 650Megs to make a backup of my system files because hard disks have a finite MTBF, viruses, etc.".

My computer repair consultant friend was telling me the vast majority of his clients have 50 CDRs of music, vid, pron but no backups of their data whatsoever. I'd guesstimate that 80% of all CD-Rs are used solely to store copyrighted music and vids. Come on people, the media is real cheap compared to tape streamers. Levy exemption can be given to schools, charities.

If levies aren't applied, then the industry will push for SSSCA on CPUs, RAM, Apps (maybe by implementing .NET-DRM by installing RIAA libraries that use encryption, and in Java (import java.DRM.memoryencryptedandprotectedMP3)) just off the top of my head. If you think this is *magically* not gonna happen then go talk to some lawyers and hear them drool on about "artist's property"... property this... property that, some lawyers that are my friends have been hostile to me for even suggesting that music isn't the artist's property they're not gonna change their minds on this. I think we all know that if DRM/SSSCA happens we'll be seeing performance drops by a factor of 10 on tomshardware, new computer will be slower than old ones for a long while. Plus the following 3 scenarios:

Badly flawed SSSCA/DRM - Makes computers slow and crash, and is useless.

Flawed/difficult-to-crack SSSCA/DRM - a hostile nation's intelligence services will come up with a way to circumvent the protection which will of course be real popular, and probably not open source into which they have implanted their own version of magic lantern trojan, ducking antivirus apps.

Virtually impossible to crack SSSCA/DRM - Code not our own any more, C and ASM no longer write to the CPU but instead a .NET-like IL or protected RAM areas only. Government can censor us, RIAA, MPAA can censor us, scientology can censor us, (insert your worst nightmare here) can censor us and bin Laden can send messages to his followers DRM-potected so no intelligence service can decrypt it.

Please people, cut the RIAA/MPAA just a little slack so that they don't bring the DOJ down on our heads, especially now. If they can take down Microsoft then they can definitely slow us down or take us down as well :-( And if you think Freenet can't be blocked then talk to those Cisco people about what you can really do with layer 4 switching.

Take flame jacket off arrrrggghhhhhh Ouch! Put flame jacket back on

Read about more about them.

the end result is what's called a "Denial of Mobility Attack, or DoM. the best solution is to firewall against such attacks. simply build a nice sturdy wall about every 3 feet or so for slipping people to catch their balance on, and prevent further mobility denials. ciscocisco is I hear working on a solution as we speak, it's been codenamed the brick and mortarist 3000 series firewall.

Cisco released IPv6 IOS images back in June with IOS 12.2(2)T. Note that this was the first commerical release, there was a earlier EFT release about for quite some time that served as a beta. The major features are there: IPv6 routing, support for stateless autoconfig, IPv6 address family support in MBGP, support for RIPng. No other routing protocols yet.

You can check out Cisco's IPv6 page for more information.

Juniper also has IPv6 available, here how to configure IPv6 on JUNOS 5.1.

Cisco now has IPv6 in Cisco IOS 12.2T and higher:

http://www.cisco.com/warp/public/732/Tech/ipv6/ipv 6_techdoc.shtml

At least in the kernel on Red Hat 7.1, they
do not enable IPv6 by default:

% ping6 ::1
socket: Address family not supported by protocol

Does Debian enable IPv6 support in their kernel
by default?

If there was a mainline Linux distro that
supported IPv6 out of the box, that would be
one way to stir up demand for it.

Most of the Linux IPv6 work is going on in Japan, not in the U.S.:

http://www.linux-ipv6.org

The Linux IPv6 work doesn't seem to be the major focus of the core Linux kernel developers these days.

• 3com - no listings (no support???)
  
• Cisco Systems
  
• HP - no listings in network equipment
  
• Juniper Networks - OS support
  
• Linux IPv6 HOWTO
  
• Lucent - interesting
  
• NetBSD IPv6 docs
  
• Windows XP - Installing IPv6
  
• RFC 2492 - "IPv6 Over ATM Networks"
  

Cisco road map (pdf) for implementation of IPv6 in IOS.

The same in html from Google.

They say that by/in 2002 (hey thats now) they will have completed implementation of all IPv6 functions in the routers.

IMHO you don't have to worry about simple users submiting to webmasters "use the Windows Media Format" for your webcast .. that's alreday taken care of major players in the media and content publishing.
To mention just one Cisco AVVID more info here . Any partner under the Cisco AVVID umbrela is using Windows Media Formats to encode and broadcast video ... not Real/MP3/ogg/quicktime even thow some of the encodings might be even better in some cases.

While we offer DSL (and allow running servers on DSL, with static IPs), many members choose to colocate 1U or 2U servers to run their own web site(s).

This approach eliminates the reliability, latency and bandwidth issues that come from locating the server in your home, at the tail end of a DSL circuit. You get the same high-availability power, cooling, and connectivity as the managed services customers in the next room, at a fraction of the cost.

The biggest difference is that unlike hosting at home via DSL, turning up the bandwidth from 384K to X megabits is simply a matter of a cutting a larger check to the association, and a simple configuration change at the gateway router.

Each member gets a subnet (usually a /29) on a VLAN dedicated to their machine(s), with hard and soft bandwidth limits courtesy of Cisco's Rate Limit IOS Commands. This ensures that no one user can eclipse another, nor steal/spoof their IP addresses.

There are two major drawbacks -- This approach isn't cheap, and hardware upgrades and related repairs take some coordination for physical access to the shared rack space.

The key? Competition. SuperNet doesn't just allow for competition, the contract requires it. In both urban and rural areas, significantly lower rates are anticipated for high-speed Internet and network access, based on the fact that the infrastructure remains open to all suppliers. SuperNet's "fiber condo" model invites investment by individual ISPs, cable companies, and other Internet and network-based businesses. In addition, a big part of the cost of providing broadband service is laying of the fiber to get it to the customers. With SuperNet contractors installing the lines, the costs of providing service logically erode.

Broadband internet is the infrastructure of this centry, like roads or railways or public water systems of previous centuries.

Well, that's it, folks. It's been nice having online radio while we could.

With these new regs, it will be the death knell of webcasting. Expect Live365 to fold within a month once these new regs take effect. Nullsoft Shoutcast and Spinner will hold on a little longer, as they are subsidized by AOL, but it too will disappear. The smaller independent stations? *poof*, as another poster put it! Considering the new fees are retroactive to 1998, if I were an online broadcaster, I'd be scrambling to dismantle my setup before they find me and send me the bill!

A shame this has to happen just when SSM, Source-Specific Multicast, was getting off the ground. Finally, an almost complete rearchitecturing of the failed Internet Multicast protocol. It addresses the two primary shortcomings of existing multicast -- address shortage and DoS attacks -- and looks like it actually could have worked.

To anyone who's watched developments in online radio technology, SSM is like nirvana. The Class D multicast address shortage is solved, by effectively using 64-bit addresses: a station's existing unicast IP address is simply concatenated with a multicast address in SSM's address range (232.x.x.x, equivalent to a big fat Class A!). And there's no central authority to go through, the station just simply chooses one of these address! This effectively gives the station the capability for 16 million channels (different SSM trees of listeners).

That's right, it's finally a tree! The many-to-many multicast model has been replaced with one-to-many. Formerly, a rogue client could simply inject data into the stream, and that data would be replicated to all other listeners. Not good. Since SSM is a tree, with the originating station at the root, this problem is solved. It will become much more difficult to "jam" a SSM station (a router close to the source would have to be hacked). With these two main problems solved, Internet multicasting would finally be good to go...!

It would have been a wonderful thing, had these new rules not been enacted. This new SSM protocol might have taken off, helping to alleviate the enourmous waste of bandwidth caused by having to repeatedly unicast the same stream to each individual listener.

Possibly the only good thing that can come out of this is more exposure for unsigned garage bands. If SSM helps to reduce the bandwidth cost of streaming, and the garage band owns their own copyrights (not a member of ASCAP/BMI/SESAC/RIAA), then it might be affordable for them to broadcast online....

Cisco have a rather plainer explanation of the impact (together with a fix for their equipment). In summary, vulnerable customer facing equipment can be made to restart, or worse, if it runs SNMP, even with some access control in place.

At least in some capacity. See for yourself.

- Freed

I noticed this last week when searching Cisco's site. The addition of the "powered by Google" snippet in the upper right hand corner of the search results threw me for a loop.

I haven't noticed much of an improvement in their search results yet - perhaps it takes time to build the link relationships index?

Cheers,
J.J.

Companies such as Cisco and the likes with a huge intranet, have been using Google for some time. Use the search engine on their main page to get the idea.

To me, it was only a matter of time until they port their technology to simpler environments (home users & smaller corporations) for a fraction of the cost.

(incidentally I searched for porn and still got 4 results back :)

1958 - TheThe integrated circuit

I bet even your toaster has an IC in it :) (and don't say "how useless is that?", because my toaster makes some pretty good toast exactly how I like it every time). Most of your modern conveniences owe their existence to this fantastic idea. One thing you're failing to mention are advances on those pre-1946 concepts (IC is an extrapolation on the transistor).

1983 - PCR

It's rather difficult to study genetics without large quantities of DNA. Along with this tech there is gel electrophoresis that you use to analyze your strands.

1984 - RouterNeed I say more? Would we be on the 'Net right now without it?

There is just too much to list, but these were some of my favorites. You get the picture. There are some others, like the Operating System, where would you be without that? Or the C Programming Language as another, or high level languages in general.

I think an important point here is that the presentation talked about advances in areas where there's not much room for advancement. Trying to make any major improvements in "word processor technology" is pretty futile.

Mike

It's a concept that has existed in Frame Relay/ATM, for example, for a decade (at least on StrataCom/Cisco) equipment. They use an algorithm called ForeSight(tm) on their core switches to throttle VC traffic in the case of congestion at the source. This later evolved into the ATM ABR standard, with input from other vendors.

In this type of automatic "biological" response, as long as no other traffic is attempting to use the bandwidth, the activity is permitted (who knows, it could be a "normal burst"). When other traffic is active, the offender is throttled back to to the source to its minimum rate.

While it doesn't stop the problem, it makes the offender ineffective at impacting service. As a result, it's no longer a "denial of service". Some more information on ForeSight and ABR in this whitepaper. The functionality predates the BPX product mentioned in the whitepaper (the StrataCom IPX had it), but that's before Cisco purchased StrataCom.

Frame and ATM are "session oriented"; a PVC or SVC defined the communications along a path, so it's easy to define the parameters to control traffic characteristics. I'm not that familiar with IP QOS; is there an equivalent functionality that would apply? If so, could the problem be solved by making the attack "unattractive" (nondisruptive)?

A number of companies are pitching products that measure the characteristics of IP traffic over a network link using a variety of different metrics and solutions. CQOS, Brix, and Cisco, all have solutiuons that do this their own way. Disclaimer, I work for the first company on that list, but there is some interesting information regarding IP measurement at all those sites. I'm not sure about Brix and Cisco's products, but I know CQOS's measures down to microsecond accuracy.

There is a very interesting article about TCP that has more details on this. It's worth reading the whole article.

For those who don't have time/patience to read the full article, here's the most relevant part:

"Satellite-based services pose a set of unique issues to the network designer. Most notably, these issues include delay, bit errors, and bandwidth.

When using a satellite path, there is an inherent delay in the delivery of a packet due to signal propagation times related to the altitude of communications satellites. Geo-stationary orbit spacecraft are located at an altitude of some 36,000 km, and the propagation time for a signal to pass from an earth station directly below the satellite to the satellite and back is 239.6 ms. If the earth station is located at the edge of the satellite view area, this propagation time extends to 279.0 ms. In terms of a round trip that uses the satellite path in both directions, the RTT of a satellite hop is between 480 and 560 ms.

The strength of a radio signal falls in proportion to the square of the distance traveled. For a satellite link, the signal propagation distance is large, so the signal becomes weak before reaching its destination, resulting in a poor signal-to-noise ratio. Typical BERs for a satellite link today are on the order of 1 error per 10 million bits (1 ¥ 10-7). Forward error correction (FEC) coding can be added to satellite services to reduce this error rate, at the cost of some reduction in available bandwidth and an increase in latency due to the coding delay. "

Sorry that link was bad. here is the actual link

Being the Cisco bigot that I am, I would go with the Cisco route (no pun intended).

Cisco offers the Cisco Security Specialist 1 certification, and the Cisco Certified Internetwork Professional (CCIP) with a Security elective.

Just my dos centavos.

Being the Cisco bigot that I am, I would go with the Cisco route (no pun intended).

Cisco offers the Cisco Security Specialist 1 certification, and the Cisco Certified Internetwork Professional (CCIP) with a Security elective.

Just my dos centavos.

Being the Cisco bigot that I am, I would go with the Cisco route (no pun intended).

Cisco offers the Cisco Security Specialist 1 certification, and the Cisco Certified Internetwork Professional (CCIP) with a Security elective.

Just my dos centavos.

You can get a TCP load balancer like Cisco LocalDirector or one its competing clones. They are expensive tho ($20,000)

I think your experience and what you have on your resume is what matters most, but if you're looking for something to get you in the door as an entry level network-security guy - you might want to check out one of the various cisco certifications. Might not be specifically catered to security, but as most will tell you - setting the network up properly is the first step in securing the beast. Plus, cisco certs still hold up fairly well with managerial types and big companies.

now you have a static where internally 10.x.x.x is the ip of the router, but the outside world can see it as 12.34.56.78

a quick google search brought up this -- http://www.cisco.com/warp/public/556/9.html

See http://www.cisco.com/warp/public/701/20.html for some more information.

Actually, it's 10Mbps up to 4000 ft but that's just an estimate and it depends on the quality of cabling etc. I've briefly tried LRE at 15Mbps over 1430m (4691 ft) of old telephone cabling and it seemed to work just fine.

Cisco LRE Rates and distancies from their white papers :

5-Mbps symmetric rate (up to 5,000 feet)
10-Mbps symmetric rate (up to 4,000 feet)
15-Mbps symmetric rate (up to 3,500 feet)

Well, it actually only works over real wire (not tin cans and string or lamp cord or suchlike).. But it doesnt require Cat 5 or even Cat 3 - it will probably work over whatever crappy wire that is already installed for phone use..

And this is primarily for MDU's - eg apartment buildings, office buildings.. this tech has nothing to do with residential access.

But the products do exist..

Here are the specific Cisco products involved:

WS-C2924-LRE-XL
PS-1M-LRE-48
CISCO575-LRE

And a link that gives the specifics:

http://www.cisco.com/univercd/cc/td/doc/pcat/lre.h tm

http://www.cisco.com/univercd/cc/td/doc/pcat/lre.h tm

(There is no space between the h and the t at the end.. I have no idea why /. is doing that, but its quite annoying)

LRE never was geared as a last mile replacement. It's main market is for Hotel, Apartment, Office, etc. This article is pure marketing fluff. One thing to note, if you're not using a PBX, then you have to use a non-homologated POTS spliter. Currently cisco only has approved two at this point in time. This isn't a new announcemnt, the LRE line has been on the market from Cisco for awhile now, with the BBSM coming from the acquistion of CAIS Software.

LRE never was geared as a last mile replacement. It's main market is for Hotel, Apartment, Office, etc. This article is pure marketing fluff. One thing to note, if you're not using a PBX, then you have to use a non-homologated POTS spliter. Currently cisco only has approved two at this point in time. This isn't a new announcemnt, the LRE line has been on the market from Cisco for awhile now, with the BBSM coming from the acquistion of CAIS Software.

Quick Google search results:

I do not believe Cisco does, or ever has, positioned LRE as a "last-mile" technology. LRE is more about leveraging existing cabling infrastucture in a multi-unit facilities such as hotels and hospitals.

Cisco's LRE product offering requires two pieces:
1. An LRE-capable switch at the head-end (such as a 2900XL LRE), which terminates the LRE and has a standard Ethernet handoff to your normal data equipment. In an intergrated voice/data setup (where you're reusing existing voice cabling to carry voice AND data) you would then use their LRE 48 POTS Splitter at the head-end and hand off to the PBX before bringing everything in to the 2900XL LRE.
2. Cisco 575 CPE, which uplinks to the head-end and splits off the voice and the data. Very similar to Cisco's 600 series.

Sound like DSL? It essentially is, just on a smaller scale (3500XL/2900XL LRE costs a whole hell of a lot less then a carrier-class DSLAM). In fact, scanning over the Cisco 575 CPE Overview, Cisco declares the technology to be "based on VDSL".

Draw your own conclusions, but I have never heard this positioned as a last-mile replacement. The article never seems to hint at it either, but simply reiterate their marketing the product line for multi-tenant facilities.

I do not believe Cisco does, or ever has, positioned LRE as a "last-mile" technology. LRE is more about leveraging existing cabling infrastucture in a multi-unit facilities such as hotels and hospitals.

Cisco's LRE product offering requires two pieces:
1. An LRE-capable switch at the head-end (such as a 2900XL LRE), which terminates the LRE and has a standard Ethernet handoff to your normal data equipment. In an intergrated voice/data setup (where you're reusing existing voice cabling to carry voice AND data) you would then use their LRE 48 POTS Splitter at the head-end and hand off to the PBX before bringing everything in to the 2900XL LRE.
2. Cisco 575 CPE, which uplinks to the head-end and splits off the voice and the data. Very similar to Cisco's 600 series.

Sound like DSL? It essentially is, just on a smaller scale (3500XL/2900XL LRE costs a whole hell of a lot less then a carrier-class DSLAM). In fact, scanning over the Cisco 575 CPE Overview, Cisco declares the technology to be "based on VDSL".

Draw your own conclusions, but I have never heard this positioned as a last-mile replacement. The article never seems to hint at it either, but simply reiterate their marketing the product line for multi-tenant facilities.

I do not believe Cisco does, or ever has, positioned LRE as a "last-mile" technology. LRE is more about leveraging existing cabling infrastucture in a multi-unit facilities such as hotels and hospitals.

Cisco's LRE product offering requires two pieces:
1. An LRE-capable switch at the head-end (such as a 2900XL LRE), which terminates the LRE and has a standard Ethernet handoff to your normal data equipment. In an intergrated voice/data setup (where you're reusing existing voice cabling to carry voice AND data) you would then use their LRE 48 POTS Splitter at the head-end and hand off to the PBX before bringing everything in to the 2900XL LRE.
2. Cisco 575 CPE, which uplinks to the head-end and splits off the voice and the data. Very similar to Cisco's 600 series.

Sound like DSL? It essentially is, just on a smaller scale (3500XL/2900XL LRE costs a whole hell of a lot less then a carrier-class DSLAM). In fact, scanning over the Cisco 575 CPE Overview, Cisco declares the technology to be "based on VDSL".

Draw your own conclusions, but I have never heard this positioned as a last-mile replacement. The article never seems to hint at it either, but simply reiterate their marketing the product line for multi-tenant facilities.

I do not believe Cisco does, or ever has, positioned LRE as a "last-mile" technology. LRE is more about leveraging existing cabling infrastucture in a multi-unit facilities such as hotels and hospitals.

Cisco's LRE product offering requires two pieces:
1. An LRE-capable switch at the head-end (such as a 2900XL LRE), which terminates the LRE and has a standard Ethernet handoff to your normal data equipment. In an intergrated voice/data setup (where you're reusing existing voice cabling to carry voice AND data) you would then use their LRE 48 POTS Splitter at the head-end and hand off to the PBX before bringing everything in to the 2900XL LRE.
2. Cisco 575 CPE, which uplinks to the head-end and splits off the voice and the data. Very similar to Cisco's 600 series.

Sound like DSL? It essentially is, just on a smaller scale (3500XL/2900XL LRE costs a whole hell of a lot less then a carrier-class DSLAM). In fact, scanning over the Cisco 575 CPE Overview, Cisco declares the technology to be "based on VDSL".

Draw your own conclusions, but I have never heard this positioned as a last-mile replacement. The article never seems to hint at it either, but simply reiterate their marketing the product line for multi-tenant facilities.

----

If you'd like to read the whole faq check it out at http://www.cisco.com/warp/public/102/wlan/radio-fa q.html.

Funny. This says 6,909, with 3,070 in the US. That's not many, especially considering that they're (necessarily) clustered around population centers. The test is one day, but the failure rate is the same. 5 people take the test every day, but less than one passes.

There will never be a CCIE flood in the market. Once the number approaches n, Cisco will make the test harder. It's in their best interests.

So I thought I should do the right thing and let the owner of the offending source domain/IP know. Sent an email with details of the source IP date/time etc. The common ones were coming off big name companies.

Response? I got one return email asking for my logs.

Just out of interest, I did the same thing with some spam recently - notifying the webmaster/abuse and the owner of the source IP from whois. I sent the relevant details, date/time content of the message.

Very similar result. One reply. Telling me to send the same info to another abuse address at a different domain that is run by the same company.

What frustrates the hell out of me is that they expect me to do their goddam job for them. I gave them THEIR source IP/host/mailing details and dates/times! I'm trying to do the right thing, but do they really expect me to get the name and phone number of the offender?

Obviously, they're not interested. In their eyes, it's not the black hats, or spammers, but *I* am the problem.

Frankly, after going out of my way to help these clowns, makes me want to join the script kiddies...

Sliding windows is flow control, not error recovery.

I would trust UDP with anything I would trust TCP with as long as the application does the error checking on the data, which is exactly what they are saying their product does. TCP is really high overhead compared to UDP, and not always necessary. One of the reasons for TCP was so that programers wouldn't have to deal with as much, but if you can make something that handles it more efficiently then you only have to send a retransmit request whenever there is lost data, and not after every window.

Maybe it's my tendacy to fight for the underdog but I feel UDP has gotten the shaft. It's a great way to slam traffic around, and as secure as your application is written to make it.

Nice little doc over TCP and sliding windows for anyone that might want one.

...regarding where the Internet might take us in the next 20 years.
More information can be found here.
-k

The problem is hubs. I have yet to see a good gigabit hub for under $2k or so. Most of the gigabit-compatible hubs offered use gigabit for uplink, and a handful of 100-base-T links for the rest of the ports.

Exactly my thought. Which is why I suggested multiple gigabit PCI cards. What I didn't state was that I was assuming that this would not be unsaturated. I've seen some less expensive switches in the "off brands" like D-Link, Netgear, etc., in the sub $1000 range but we are talking about a home network.

And your internal routers are either not routing (saturated) gigabit traffic through multiple cards, or not running on commodity hardware.

True. I'm not routing gigabit Ethernet, much less saturated. The network isn't saturated, despite lots of data flowing. They are running on commodity hardware though: some rackmounts w/ oem motherboards, "obsolete" name brand "business" grade and some highend clones. Just routing packets. It takes a lot of packets to saturate. However, Cisco uses the PCI bus for their routers as well like on the 7200 or the 2600's. Pretty much all their gear. Which clearly are not comodity equipment but they are PCI.

More than one saturated gigabit ethernet interface would certainly swamp a 32-bit 33 MHz PCI bus.

I haven't witnessed it myself but you probably are right about saturation with gigabit Ethernet. I don't have a sense about gigabit Ethernet on 64-bit PCI slot.

I am a-priori assuming an application that will saturate gigabit ethernet channels. Otherwise there's no reason to use gigabit at all, as you point out.

MAE East up until a couple of years ago was using ancient (assuming an Internet generation is 6 months and a calendar generation is 20 years) by Internet time DEC Gigaswitches, which are fast Ethernet switches. Designed to burst up to just over a GB for the whole switch, my recollection is they were operating at about 1.6 GB sustained in late 1997 or early 1998. This was basically saturated so you can imagine my wondering about saturating gigabit Ethernet at home.

I find given current needs in a home that gigabit Ethernet would be needed on the mind-boggling side except in rare instances. Sorry for the crack about crack ;-) I think you and I are largely on the same page as it were.

The problem is hubs. I have yet to see a good gigabit hub for under $2k or so. Most of the gigabit-compatible hubs offered use gigabit for uplink, and a handful of 100-base-T links for the rest of the ports.

Exactly my thought. Which is why I suggested multiple gigabit PCI cards. What I didn't state was that I was assuming that this would not be unsaturated. I've seen some less expensive switches in the "off brands" like D-Link, Netgear, etc., in the sub $1000 range but we are talking about a home network.

And your internal routers are either not routing (saturated) gigabit traffic through multiple cards, or not running on commodity hardware.

True. I'm not routing gigabit Ethernet, much less saturated. The network isn't saturated, despite lots of data flowing. They are running on commodity hardware though: some rackmounts w/ oem motherboards, "obsolete" name brand "business" grade and some highend clones. Just routing packets. It takes a lot of packets to saturate. However, Cisco uses the PCI bus for their routers as well like on the 7200 or the 2600's. Pretty much all their gear. Which clearly are not comodity equipment but they are PCI.

More than one saturated gigabit ethernet interface would certainly swamp a 32-bit 33 MHz PCI bus.

I haven't witnessed it myself but you probably are right about saturation with gigabit Ethernet. I don't have a sense about gigabit Ethernet on 64-bit PCI slot.

I am a-priori assuming an application that will saturate gigabit ethernet channels. Otherwise there's no reason to use gigabit at all, as you point out.

MAE East up until a couple of years ago was using ancient (assuming an Internet generation is 6 months and a calendar generation is 20 years) by Internet time DEC Gigaswitches, which are fast Ethernet switches. Designed to burst up to just over a GB for the whole switch, my recollection is they were operating at about 1.6 GB sustained in late 1997 or early 1998. This was basically saturated so you can imagine my wondering about saturating gigabit Ethernet at home.

I find given current needs in a home that gigabit Ethernet would be needed on the mind-boggling side except in rare instances. Sorry for the crack about crack ;-) I think you and I are largely on the same page as it were.

Don't you read slashdot everyday? Shame on you if you don't!

Just a few weeks ago there was this article about 3com faceplates. You can consolidate all your phone and data into 1 jack while running just 1 cable run to your rooms. Now if you got money to burn, you might consider one of these babies from Cisco. This is their media convergence server which will combine voice/data/video into something that can run over cat5. A MCS will cut out your need to run separate phone and video lines. Hey want to hire me to set this stuff up? I'm totally jobless right now and could really use the money :) Good luck on your house!

The distance will be effected by how much power the FCC will allow us to use in this frequency range. It will may vary from country to country. The IEEE Standard covers how the thing is going to communicate between vendors products (Lucent and Cisco have to play nice with each other). The bandwidth allowed will depend on how many channels we are allowed to use, or the product will let us use. One GHZ channels should be able to pump DS3 or higher speeds...Let's keep our fingers crossed. I'll be happy to beta test any gear :)