What's new

Has Networking Lost Its Mojo?

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

vnangia

Senior Member
As a newcomer, I for one would like to say no.

I've just recently started seriously looking at my network and I'm finding all sorts of things to improve around here, and learn about. Since the start of summer, I have:
  • Built a NAS.
  • Upgraded it once.
  • Converted to a gigabit network, as much as possible.
  • Optimized said gigabit network to get better performance out of it than I get at work.
  • Started building a router.

I know that at some point of time I might well get bored of this hobby. I'm a strong believer in this article's basic assertion that geeks need projects. And right now, my network is my project and I'm more excited about it than I have been about anything else in a while.
 
Well, for those who just started, they may be exciting. ( Like all things.... ) But as soons as you ran into limitation ( like us old peeps ) and have been waiting YEARS for solution. I can understand why this topic started.

Yes it is boring. Because
It is slow. Gigabit ethernet is slow. At least those integrated in the motherboard hasn't improved at all. Router and Switch cant handle multiple gigabit connections with some decent speed. I.e the low end router and switch hasn't improved at all.

NAS is slow. As recently show the bottom, home user segment aren't having any speed improvement. Although it is getting much more stable. We dont have to restart and reset our NAS every now and then.

Wireless, like Tim said, is just some money making pile of crap that is like subprime lending, it is being packaged and marketed but it simply does not work. With Buildings that have concrete walls, ( high population area ) wireless is virtually unusable. Speed is slow, even with Draft N, streaming HD video is still lagging. Reality is we are still YEARS away from true wireless.
Most of the people i know changed back to cable because they are sick and tired of Wireless performance.

I hope the promise of Ethernet Firewire will one day come. Which promise faster then current Ethernet and easier setup.
 
Bob!

The holy grail in the corporate world is the "Branch Office Box". Consider the current needs of someone working from home or an office of up to 5 people. They need VPN, WAN optimisation, security (ie protection from potentially malicious incoming traffic, health screening and protection of the corporate network from them), auditing, backups (local quick ones, deferred behind the scenes upload to central facility), remote access, print servers, peripheral hosting such as scanners, separation of home use from business use, incoming and outgoing traffic management and the list goes on.

This functionality is all currently achieved by a hodge podge of random hardware and software products, usually requiring an expert to install and configure if they have to have any semblance of working together, not to mention ongoing maintenance.

A single branch office box that does all this stuff in one unit (and does it well) is what the likes of Cisco want to come out with. It would also drive their larger central office appliances - the WAN optimization, backup, health control, vpn etc all have to connect to something on the other end.

It will be interesting to see of the open source space comes out with something first. You still currently need a hodge podge of random open source products and an expert to configure and maintain. Two players to watch are Vyatta and Untangled although neither is going anywhere near storage and backups.
 
firmware

I have to admit the thing that's keeping me going is open source - my NSLU2 is a full fledged server now, and my wrt54g is running tomato, doing QOS and giving me pretty charts of the bandwidth out of my house. I do wish we were seeing higher throughput at the home network / SOHO level, but frankly, until internet speeds reach those of all the other developed nations, I'm not sure how much impetus there will be for it.

I'd love to see more developments on the security front for wireless - I'd love to have simple, key-free (well, free from me or any of my guests having to type anything in) encryption. And I'd like to see more interesting developments with music servers - I feel like right now it's something of a hodge-podge.

I did like the idea of the Belkin N1 Vision. I hope to see more like it (but with good firmware, of course).
 
Just need some imagination :)

Yes, basic SOHO net stuff is becoming mainstream - most stuff just works and most SOHO users can fumble their way without being uber geeks anymore.

This is true of SOHO and PC tech in general - check out other sites like Toms Hardware or Anandtech - they used to be cool places where everyone was excited about new processor tech, GPUs, high speed interconnects, etc. etc. Now they are just lists of CPU charts, video cards, and mind numbing articles comparing 1TB drives with 1.5Tb drives. Any idiot can overclock nowadays and the real techies don't even bother. The excitement about SSDs was over before it even began. Its a bit like comparing CAT5e with CAT6 - is it really worth writing about?

There is still more than ever geeky stuff out there: lock/wait free algos, multicpu/multicore programming, high speed interconnects (i.e. Infiniband), etc. but its all too geeky for general interest. For the mainstream single socket PCs hooked up to GigE with a basic RAID1 NAS is more than good enough tech.

If you want to peak geek interest then choose particular SOHO scenarios and investigate geeky solutions i.e. NAS for SOHO render farms, build your own CISCO VPN concentrator using CentOS, how to implement continuous incremental real-time backups across your entire NAS cluster, cut you power bill by using automated shutdown and timed wake-on-lan across all your networked devices, investigate opensource network applications and what you can do with them (rsync, netcat, tic, emcast, ssh, vtun, vlan, flyback, GlusterFS, etc etc), real-time SOHO system monitoring with flashing red lights on a big wall mounted monitor, or jump off the deep end with things like OS bypass to minimize latency for custom real-time applications (i.e. GAMMA, libe1000, etc.).

If you have $$$ or can get companies to lend you products definitely check out hardware accelerated devices - pretty much anything you can find will have some geeky purpose (NetVCR, Infinistreams, Exegy, Solace Systems, KillerNIC, Voltaire switches, anything with a FPGA, NICs with SBCs onboard, etc.)

Keep covering the mainstream products but branch off into geeky projects - the more geeky the better - that's where the magic is - every SOHO dreams of being an Enterprise :)

PG
 
Tim!

Don't throw in the towel - your site is really starting to establish itself as a great authority for comparing all net devices in our range. I've never found anything quite so clear at cross-product comparisons as your Router Charts and NAS charts. They really are quite irreplaceable!

Things that I find quite interesting and fascinating in the network space these days include:
  • PoE
  • VoIP and Unified Messaging with Asterisk
  • High Performance NASes - topping your charts
  • MoCA - where GigE just can't go
  • uPnP and DLNA - still searching for that killer setup
  • Transcoding farm and Render farm
  • iSCSI and AoE
  • The Cell Processor - How to harness for Render or Transcode farm
  • RFID hacking - while it's still legal
  • LAN-to-LAN VPNs - IPSec or SSL
  • OSPF - dynamic routing
  • In-Car Network and "Auto Sync when I enter the garage"
  • Diskless network booting with gPXE
  • ZFS - when will we have it in a NAS?
  • Coda or AFS or INtermezzo - true distributed Filesystems
  • Rsync and Unison - the Sync Holy Grail
  • Safe "Cloud Computing" - Jungle Disk? ADrive?
  • Tor and other anonymizers
  • Nagios and Cacti and MRTG - and other Open Source Network Monitors
  • Vmware Server, ESXi, Xen, KVM
  • Anything that uses ffmpeg
Not exactly sure how much of this falls inside the purview of your site, but there's gotta be enough in there to whet an appetite or two....
 
You know Tim, I've been working in or around the IT sector for over 15 years now and I'll admit that I still get fired up about things every now and then. What I am tired of is the whole "we'll sell you a feature set...and now you can debug it for us" mentality which is exactly why Macs are so popular these days. Certainly I'm in a constant struggle between the desire to simplify life in every respect, and a curiosity about what's new in the world of technology.

What does keep me going is the standing system analyst challenge to take a complex system and make it simple to the end user, and reliable to use. So even though my work these days running Cinevate has little to do with IT, what is rewarding is to tackle the challenges of my own (and a few other businesses) systems to add all those features Roger mentioned in post #3. One of the business owner's I still help out was tickled pink that his new wireless laptop could connect to the web while he was traveling and give him access via VPN to his office desktop. We use a combination of VOIP, WIFI, WIFI VOIP phones, VPN, and as you know, have been experimenting with NAS units for video editing. One potential objective is to then provide the "office package" for others not quite so tech savvy in our marketplace. A good example of simplifying is the whole media streaming thing. Virtually nothing out there decodes and plays everyone's files perfectly. So why bother trying? Using a simple HTPC eliminates this by just using VLC and a codec pack or two. I've bypassed this altogether again by just using baluns that take component HD and digital audio over CAT5 to my home theatre system. Using a bluetooth keyboard/mouse which lives in the home theatre...now there's no HTPC, no media streaming and no pains in the butt. The HDTV in the home theatre is just set up as a third monitor off my editing workstation.

This leads to sites like this and how useful they are. Check your web stats (as I know you must already do) and take a look at how many folks just come here, read what they need and then wander off. Objective information on the web is increasingly hard to find because many reviews are written by folks just trying to sell something, or make an advertiser happy. Consider how successful the Consumer Reports concept is, and then compare it to what you do here :) I'm with Roger though in that I'm still searching (and getting closer all the time) to this goal of an office that has all the connectivity it needs, complete data security, and reliable systems in place that make it all happen. The picture gets even more texture when you consider that our entire business runs from a website based on 3 open source software packages, and does this for a ridiculously small yearly fee. We recently started using web-based tools to closely collaborate with a few of our manufacturers. Yes life is short, and everything we surround ourselves with tends to break and suck more time out of our brief visit here on the planet. Still, it's pretty amazing what you can do with some time, a problem solving ethic...and information provided on sites just like this one :)
 
Last edited:
Hell yeah ...

make it faster. I am using computers for 20 and networks for about 15 years now ... old BNC stuff and the days where 10MBit made you wow! ... my first harddisk was a 20 Megabyte external drive for my Amiga 500 ... it was loud as hell and I thought I would never fill it ... Boy was I wrong.

Now my home DSL bandwidth is 50/10 MBit, my CPU is a dual core monster going to be upgraded to a even bigger quad core monster to be able to munge away the gigs and gigs of more or less nice 12 megapixel raw images I shoot with my DSLR.

CPU power? No more exiting ... just keeps getting faster. Moore was right 'til now and honestly, I am waiting for the day he is proven wrong - either direction, I don't mind. GPU power? same ... hard disk speed? laaaaaaaaaaame. SSD? ok ... a logical step ahead, what's next. Ok, make them really energy saving and really fast but that's around the corner with Intel SSDs and more coming up. 1.5TB? Snore ... perpendicular recording? Made me listen up a second.

This is where it starts to get interesting in regards of personal need and, of course, this topic: Where do I put my gigs and gigs I mentioned above? Save my digital content? Share it with my wife? On my sofa?

I started looking for a save storage solution a year ago or so? Did not find any which suits my needs. Capacity is not a problem but speed is. A gigabit NIC should be able to perform with somewhere around 100MByte/s between clients. SOHO equipment does not even come close. Or do I get something wrong here?

Now I am looking for DAS solution. Because I want to safe my stuff in first place. So I'd like to see more tests of DAS equipment for the SOHO users. SOlutions which let me backup via a fast eSATA to an external drive/RAID

For sharing I am still looking - and waiting. It is not that important to me. But for many others and the market is huge! Hell, with my internet bandwidth I have the feeling that I don't need NAS anymore sometimes. I just upload to the internet from my home office and download with a wireless connection in my living room ... ok ... just kidding as local NAS is sure faster with uploads. But with some devices I doubt it on the download ...

To conclude my lengthy posting: Make it faster.
 
I don't know about you, but I'm pretty down these days about the state of things in SOHO / SMB Networking land. I mean, when was the last time that you were excited about a networking product? For me, it's been a long while.

Tim, don't lose heart. The area is actually heating up a great deal, but these things take time.

E.g. The next TV I'm going to own will have an Ethernet port. My Blu-ray player already has one, and I've already used it for an automated firmware upgrade.

E.g. Really high-speed broadband is just taking off. This is a long-haul development, which takes time and money, but when it does take off, it's going to give another big push to home networking.

Networking should be stuff which just works and you don't have to think much about. As such it isn't meant to be "exciting", but help you get things done. For the most part, it's lived up to this sort of role, and role is still expanding significantly.

Cheers.
 
I would agree, to an extent, that home networking has been a little boring as of late. Wireless routers are standard, NAS's are commonplace, streaming is no longer a task.

Personally, I never find myself bored with networking, especially the more I delve into my Cisco curriculum (CCNA/CCNP). There's always something new I'm learning. I had a chat with a really smart CCIE guy recently, and he's forgotten more about networking than I could ever dream to learn. For me, networking hasn't lost its mojo, but it's definately leveled off in the home space a bit. Some of it I think depends on your knowledge and skill level. Whenever I help friend set up networks, they're amazed as to the kinds of details involved. When I started out doing networking, NAT was something that was really cool, now it's like 'meh'. But the more I learn about NAT and PAT and the differwent technologies (via. CCNP), the more interesting it becomes again.

It'll get interesting again once some new technologies come online. I think we've just recently passed through, and we're on the tail end of the first networking 'boom'. Another one will come along.
 
Thanks, everyone, for the thoughtful posts. One of the things that has kept me doing this for 10+ years is that I thought that there were people like you out there. It's nice to know that there really are, and that at least some of them take some of their valuable time and spend it at SNB.

The article wasn't meant to say that I am throwing in the towel. But more to vent and get a discussion like this going. Looking back at your posts, I think there is one comment that states exactly what I'm feeling:

Dennis' comment:"Certainly I'm in a constant struggle between the desire to simplify life in every respect, and a curiosity about what's new in the world of technology."

One of the reasons that SNB exists is that networking still doesn't "just work" for people who just want to get on with it. They don't want to futz with settings and tweaking and having the latest and greatest. They just want all of their connectable gizmos to, well, just connect!

In some cases, like the closed, tightly-controlled Apple/MacOS/iPod Stevie's world, it does. But that ease of use comes at the price of limited choice.

In the larger Windows dominated world, Vista didn't really advance the ball and make things easier and more reliable. Instead, it attempted to hide information and controls that, due to the nature and history of Windows, shouldn't be hidden. And made too many people have to deal with the full-frontal complexity of Windows server file permissions to get a simple protected share running.

SNB does get torn in two directions. One is represented by you guys; advanced, knowledgeable practitioners in the ways of the network. I don't want you to get bored and go away. To the contrary, I want you more engaged and to actually participate more by writing about many of the topics that you have suggested (BTW, that's a great list, corndog! When can I see the first piece?)

The other direction is to be an advocate and protector of the average person whose eyes glaze over when he or she walks into the Networking section at Best Buy and tries to decide what to buy. Like the good engineer that I am, I try to take a data-driven approach. I put the performance data in the Charts, write detailed reviews, provide Slideshows of user interfaces and supply how-to's for buying routers, NASes and wireless.

But, in reality, many of these folks just want to be told what to buy, whether or not it's really suited to what they actually need. And most of them never make their way to SNB in the first place, because we don't have the reach that the larger sites and print publications have.

The main point of the piece, however, is the lack of innovation that I see in Networking. Most products are minor tweaks on the same reference design provided by the chipset manufacturer, with the only value-added coming in the form of a prettier user interface or setup wizard. Too often, the products are still flaky in operation, hard to configure and don't "just work" when installed.

Where are the NASes that have the backup intelligence built into them, instead of requiring an app to be installed on each client? And why can't that NAS scan the network for shares and guide the user in setting up a scheduled backup for each of them, that then runs reliably and communicates its status when and how the user desires?

Where are the wireless routers that don't make the user choose an operating band, but instead handle the decision based on traffic type and the surrounding RF environment, changing channel and band as needed?

And why aren't the vendors getting the word out on much easier to use SSL VPN's and integrating it and no-brainer remote access into their "consumer" product lines? And while they are at it having Vista compatible SSL applets in place when they ship product?

No, folks, the truth is that, aside from Cisco, who seems entirely uninterested in innovating in the SOHO market, none of these companies have the R&D money to spend on real innovation. So, as far as I can see, more of the same will continue to be the story in consumer networking.

Now, excuse me, I have to go write a review on a product that is a perfect example of this topic.
 
It's the clients

One of the things I think that is making it somewhat boring is that the clients a fundamentally the same.
Networking has always been about the exciting things that you can achieve when you connect. The networked home becomes exciting when you consider the possibilities, but is boring when you look at the reality.
The reality is that the networked home is still a PC centric get to the internet environment, and it has got pretty boring. There are some home entertainment/ automation / lifestyle solutions out there but they are still too expensive, too proprietary, and to clunky for mainstream use.

If we can shake up the client environment things will get more exciting
 
Plastic optic fiber, maybe ?

I think it was no mentioned yet, I hope POF will bring back some appeal to wired networking. Non obstrusive, fast enough, can go into electrical wiring conduits.
 
I think it was no mentioned yet, I hope POF will bring back some appeal to wired networking. Non obstrusive, fast enough, can go into electrical wiring conduits.
Netgear talked about this at CES in January. But, like MoCA coax adapters, still not shipping.
 
Another thing that I've watched for a while, hoping it would finally turn into something real, is HANA. This is a Multimedia networking standard based on firewire.

I've always been a fan of firewire, for one reason: isochronous transfer. Every other networking technology that SOHO people use (ethernet, WiFi, MoCA, HPNA, powerline, phoneline, etc) on a daily basis is designed around the "best effort" delivery approach. Of all these various options, ethernet is the ruling titan, but that's just because it is the fastest and most reliable. But even ethernet is "best effort". Firewire is different. Isochronous transfer is a dedicated, guaranteed bandwidth technology, which sets it completely apart from all other types of networking.

HANA seeks to leverage this "special" characteristic of firewire, and put it to work carrying HD audio/video in multiple simultaneous streams. And with the recent developments in firewire, this is easily possible.

Everyone is talking about the death of firewire. I for one hope it doesn't happen. Currently, firewire drives are still the better external option, and pro audio is starting to really put firewire to work with the mLAN standard. The new F22 Raptor's fly-by-wire system and internal monitoring and information systems are all firewire based. If HANA can reduce the cable clutter in the average home theatre setup to a simple firewire cable, I think it will be here for a long time.

Then it would be great to see some real FW1600 and FW3200 products come out.

Yeah, I'm an optimist.
 
The main point of the piece, however, is the lack of innovation that I see in Networking. Most products are minor tweaks on the same reference design provided by the chipset manufacturer, with the only value-added coming in the form of a prettier user interface or setup wizard. Too often, the products are still flaky in operation, hard to configure and don't "just work" when installed.

Where are the NASes that have the backup intelligence built into them, instead of requiring an app to be installed on each client? And why can't that NAS scan the network for shares and guide the user in setting up a scheduled backup for each of them, that then runs reliably and communicates its status when and how the user desires?

Where are the wireless routers that don't make the user choose an operating band, but instead handle the decision based on traffic type and the surrounding RF environment, changing channel and band as needed?

> Most products are minor tweaks on the same reference design provided by the chipset

Because the money is in the volume sales and R&D costs. And unique things have extra support costs. So you see the under $100 market at BestBuy and the over $1000 market for people who can afford monthly support contracts. The range in the middle is hard to target. (But I'll ask some question about one of these devices in a new post in a minute.)

> backup intelligence built into them, instead of requiring an app to be installed on each client?

My clients are architects. They are not backing up a few word documents to an external drive. They backup 1 to 10 gigs a day and need to be able to back track to versions from YEARS ago. And this is a small office of 25 people. The NAS making choices would either require way too much time, bandwidth or not give us the archive granularity we need. But I have to think that Apple's Time Capsule is a big start. But it does have the drawbacks you mentioned about the Mac. And I really like supporting Macs. :)

> wireless routers that ... handle the decision ... changing channel and band as needed

Again Apple wireless routers do some of this now. But this is a really hard thing to program into a small "lump".

But the main issue is sale price. It drives the market. People will buy an iPhone at $200 with a $150 month fee but not buy it at $400 with a $100 per month fee. Go figure.

David
 
Another thing that I've watched for a while, hoping it would finally turn into something real, is HANA. This is a Multimedia networking standard based on firewire.
I have visited the HANA booth at CES for the past few years. Interesting technology, but it doesn't seem like it is going anywhere in terms of widespread adoption as a home media distribution method. But neither is anything else.
 
But the main issue is sale price. It drives the market. People will buy an iPhone at $200 with a $150 month fee but not buy it at $400 with a $100 per month fee. Go figure.

I'm afraid you're right, David. Purchase price seems to be the driving factor. But Apple continues to show that at least some people will pay a higher price for ease of use.

I also take your point about the semiconductor cost/volume curve. But I see a lot of the problems as being in the software/firmware, not silicon.
 
Yep, life was supposed to be simpler right now with folks wandering around trying to figure out what to do with all of their leisure time :) Funny how it worked out.

I think the big winners in the market will be the enterprises like Apple that spend thousands of hours figuring out how to remove menu options, user options etc. in the interest of just having a power button and systems that work seamlessly at the human interface side of things. With technology moving so quickly, it takes a very special company and a dedicated team that cares enough to do it.
 
I have to admit, yes it has gotten a bit boring.

I got excited a few years ago with the explosion of the broadband market, and the line of routers that came out due to it. Aside from growing a bit in horsepower, routers haven't really changed much lately.

The evolution of UTM appliances brought a breath of fresh air and excitement to this part of networking. I believe the demand for these will grow substantially.

Remote Access was exciting a few years ago, with basics such as VPN, remote desktop, or even pre-remote desktop technologies such as PcAnywhere, *VNC, Timbuktu, CarbonCopy, or fancier setups such as the Remote Web Workplace of SBS, the demand for setting these up has been squashed with the release of free remote desktop access programs such as logmein.

NAS hasn't really excited me at all. The way I look at it, with the evolution of hard drives hitting the size of the state of Montana, and computers commonly having built in SATA RAID, they just haven't sparked my interest.

I've been absent in the area of media servers....this is one area perhaps I could get interested in, and get a flutter back in my eye.
 

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top