atcomsystems.ca/forum Forums VOIP Phones & IP PBX Networking Gigabit to Desktop: Is it a hype or real demand?

I took this post off topic, so I'll take it right back to the original post. You are asking if gigabit is hype from equipment vendors, or if there is real demand from users. The key here is users, and not applications. I can argue about all of the real world applications that I see every day that create a demand for gigabit. You in turn can argue that we all need to have a cup of coffee and transfer small files. The bottom line is that we're talking about applications for gigabit, and you specifically asked about demand from users. So, do users want gigabit?

Yes, in my experience there absolutely is a demand from users for networks that are at least capable of running gigabit. That doesn't mean they have applications that require gigabit, but this is what users want. We have had threads here which discuss clients wanting Cat6 or Cat6A installed instead of Cat5e. They want the higher category data cable to support gigabit and beyond. Customers are willing to pay extra for Cat6 even when we tell them Cat5e is fully capable of running gigabit. The majority of users do not want to have an infrastructure that will limit them. The last line of the article you linked speaks to this:

"For a few bucks more, why not future-proof your network now?"

Again, this article is from 2002, and gigabit was around 5 years before then. If you want to future-proof your network now you install a cable plant capable of 10gig. Wait 5 years and the majority of users will want their network to be capable of 10gig. They may not need it, they may not use it, but that doesn't mean the user demand won't be there.

This discussion reminds me of the arguments I heard (and made) when telephones went from 25pr 1A2 to "skinny wire" electronic... or from 2pr. analog to 1pr. digital... or from TDM to VoIP...

The fact of the matter is that staying in business is all about customer demand and customer demand isn't driven by the customer's understanding of technical details, by the demands of the their application, or even common sense assessment of ROI... it's driven by customer perceptions. Research, manufacturing, marketing... all those things push technology faster and bigger. Then it's like climbing Mt. Everest... they want it because it's available.

If customers feel they need newer technology, that's what they'll demand, and that's where the market will go... with or without us. We can talk ourselves blue in the face about why things shouldn't change and we can be absolutely right. Meanwhile, somebody else is responding to that demand and eroding our marketshare.

I would have to agree with Clinton partially. when we wired our new building in April the only reason we didn't go 10G was cable cost. We did put conduit in all the walls, and the new space is wired with 6, and not 6a. We left room to add 6a if a particular client project needs it.

Part of it is user education where more internal speed wont help their internet access, however we are moving more and more data across the network. from a bandwidth standpoint. IP phones, Video Conferencing, video based training, security cameras, hard drive images (and yes those goto end user systems)

John, the 140 Gig is created as a single file
across the network. Not created local and copied. True, I/O bus on either system can limit the data transfered, and it is the case right now, but the 2 hour difference in backup time was worth the change. Backup including verify is currently taking 12 hours, and was taking 14 hours before. if I can shave another 2 hours off by upgrading the destination to faster drives and a better IO bus, that will be money well spent as well. That particular system has some other issues and is scheduled for replacement.

This selling point seems to be more as an insurance than real performance/feature advancement that average users can feel or enjoy. However, I guess this insurance for possible technology obsolescent is never as meaningful as my auto/home/health insurance.

I rarely saw any other new technology (when they were new) was marketed this way by telling people: buy it, it may be useful in future. This was not the case when people moved from mechanic typewriter to word processor, from analog PBX to VoIP, from fixed landline to mobile cellular phone, from film camera to digital camera, from tube TV to LCD or plasma TV, from dial-up modem to DSL or cable modem and so on so forth. My point is that people should feel the real impact/benefits of a new technology in a substantial way as soon as they start using it. Otherwise, it may just be a hype.

From another perspective, I would say that "gigabit to desktop" maybe more meaningful only when applications such as "HDTV to desktop" becomes a necessity for daily business communications. But even "HDTV to desktop" can be handled nicely by fast Ethernet as its bitrate should be less than 20Mbps. Probably "Multi-channel 3D HDTV to desktop" would out-run 100Mbps? This may be why Cisco's has been pursuing their Tele-presence project to create a demand for "Giagbit to desktop". But I doubt it would be at least 10 years away if it could become real for average office users?

In the same 2002 article, it says "The biggest obstacle I see to Gigabit on the desktop is that there's no killer app for it......".

Today(7 years later), there is still not a single killer application for "gigabit to desktop", and I can safely bet there would not be any either for the next 10 years.

By calculation, moving 140GB data at 1G speed would take less than half a hour, which is quite short compared to the 12-hour backup time. This is an example that the bottleneck may not be the network speed, but the IO devices, which should be true in most real world situations.

Gigabit ethernet over the powerline is here - pretty cool and not that expensive https://catalog.belkin.com/IWCatSectionView.process?Section_Id=206578

From their website: "The standard transmission rate—1000Mbps—is the ideal physical data rate. Actual data throughput and distance will be lower, depending on interference, network traffic, building materials, and other conditions."

With added overhead of doing encryption, the actual data throughput is likely well below 1G, but it is definitely higher than current 85Mbps technology.

Although 100Mbps should be fast enough to stream HD movie, "Gigabit to Entertainment center" or even "Gigabit to Home" should be more interesting for residential users than "Gigabit to desktop" for business users.