Tag Archives: DOCSIS

The Upload Crisis

Carriers continue to report on the impact of COVID-19 on their networks. One of the more interesting statistics that caught my eye was when Comcast reported that upload traffic on their network was up 33% since March 1. Comcast joins the rest of big ISPs in saying that their networks are handling the increased traffic volumes.

By ‘handling’ the volumes they mean that their networks are not crashing and shutting down. But I think there is a whole lot more to these headlines than what they are telling the public.

I want to start with an anecdote. I was talking to a client who is working at home along with her husband and two teenagers. The two adults are trying to work from home and the two kids are supposed to be online keeping up with schoolwork. Each of them needs to create a VPN to connect to their office or school servers. They are also each supposed to be connecting to Zoom or other online services for various meetings, webinars, or classes.

These functions all rely on using the upload path to the Internet. The family found out early in the crisis that their broadband connection did not provide enough upload speed to create more than one VPN at a time or to join more than one video call. This has made their time working at home into a major hassle because they are being forced to schedule and take turns using the upload link. This is not working well for any of them since the family has to prioritize the most important connections while other family members miss out on expected calls or classes.

The family’s upload connection is a choke point in the network and is seriously limiting their ability to function during the stay-at-home crisis. But the story goes beyond that. We all recall times in the past when home Internet bogged down in the evenings when everybody in the neighborhood was using broadband to watch videos or play games. Such slowdowns occurred when the download data path into the neighborhood didn’t deliver enough bandwidth to satisfy everybody’s request for broadband. When that download path hit maximum usage, everybody in the neighborhood got a degraded broadband connection. When the download path got overloaded, the network responded by giving everybody a little less bandwidth than they were requesting – and that resulted in pixelating video or websites that lose a connection.

The same thing is now happening with the upload links, but the upload path is a lot more susceptible to overload.  For technologies like coaxial cable networks or telephone DSL the upload path leaving the neighborhood is a lot smaller than the download path into the area. As an example, the upload link on a coaxial network is set to be no more than 10% of the total bandwidth allowed for the neighborhood. It takes a lot more usage to overload the download path into the neighborhood since that path is so much larger. On the upload path, the homes are now competing for a much smaller data path.

Consider the difference in the way that homes use the download path compared to the new way we’re all using uploading. On the download side, networks get busy mostly due to streaming video. Services like Netflix stay ahead of demand by downloading content that will be viewed five minutes into the future. By doing so, the neighborhood download network can have cumulative delays of as much as five minutes before the video streams collapse and stop working. The very nature of streaming creates a buffer against failure – sort of a network insurance policy.

Homes are not using the upload links in the same way. Connecting to a school server, a work server, or a video chat service creates a virtual private network (VPN) connection. A VPN connection grabs and dedicates some minimum amount of bandwidth to the user even during times when the person might not be uploading anything. A VPN carves out a small dedicated path through the upload broadband connection provided by the ISP. There is no buffer like there is with downloading of streaming video – when the upload path gets full, there’s no room for anybody else to connect.

The nearest analogy to this situation harkens back to traditional landline telephone service. We all remember times, like after 911, when you couldn’t make a phone call because all of the circuits were busy. That’s what’s happening with the increased use of VPNs. Once the upload path from the neighborhood is full of VPNs, nobody else is going to be able to grab a VPN connection until somebody ‘hangs up’.

Residential customers have historically valued download speeds over upload speeds and ISPs have configured their networks accordingly. Many technologies allow an ISP to balance the upload and download traffic, and ISPs can help upload congestion by providing a little more bandwidth on the upload stream. Unfortunately for cable companies, the current DOCSIS standards don’t allow them to provide more than 10% of bandwidth on the upload side – so their ability to balance is limited.

As I keep hearing these stories from real users I am growing less and less impressed by the big ISPs saying that everything is well and that their networks are handling the increased load. I think there are millions of households struggling due to inadequate upload speeds. It’s true, as the big ISPs are reporting, that the networks are not crashing – but the networks are not providing the connections people want to make. No big ISP is going to admit this to their stockholders – but I bet a lot of those stockholders already understand this first-hand from having troubles trying to work from home.

Cable Companies and 10 Gbps

One topic covered extensively at the recent SCTE-ISBE Cable-Tec Expo in New Orleans was the ability of cable networks to deliver 10 Gbps broadband to customers. The fact that this is even being discussed is a testament to the fact that big ISPs all acknowledge the huge growth of demand from consumer and business broadband in the country.

Most urban cable companies just made the upgrade to DOCSIS 3.1 in the last year or so that allows them to offer gigabit products to customers. Everybody acknowledges that the need for 10 Gbps products is likely at least a decade away, but now is the time to start the technology research needed to create a product in that timeframe.

It’s been clear for some time that cable companies don’t want to lose the speed battle and are working to compete against the introduction of fiber in urban markets. The number of households being passed by fiber continues to grow. AT&T built past millions of homes in the last few years, mostly in small pockets around existing fiber nodes. CenturyLink even built residential fiber for a few years before abandoning the concept to concentrate on building fiber to businesses. It’s not clear who might build urban and suburban fiber, but the fact that the cable companies are looking at 10 Gbps speeds means they think that somebody will do so.

Other than some limited cases, most fiber providers are still building fiber networks with 1 Gbps fiber speeds. Verizon is building a 10 Gbps fiber network to supply bandwidth to small cell sites but is not yet using the new technology in the FiOS network. The whole fiber industry is waiting for one of the big ISPs to embrace 10 Gbps products to help pull down equipment prices, but that doesn’t look likely to happen any time soon.

There are significant upgrades needed for the cable industry to offer 10 Gbps speeds. A 10 Gbps downstream data path requires 1.3 GHz of bandwidth, which is greater in capacity than all but a handful of cable networks. Adding a decent upload data stream and still carry TV channels means that cable systems will need to upgrade to 2 – 3 GHz of bandwidth. That’s a major upgrade and would likely require replacing most or all of the amplifiers and power taps in the outside coaxial cable network. This would also likely require some replacement of older coax cable. Upgrading to faster speeds would mean an upgrade to headends as well as to the millions of DOCSIS modems sitting in customer homes.

I’ve heard speculation that cable companies will consider an upgrade to fiber rather than going to 10 Gbps over DOCSIS. Almost every cable company is now using PON technology when building to greenfield subdivisions. While it’s expensive to build fiber to every home, almost every CEO of the big cable company has acknowledged that their eventual future is with fiber. Altice is already pursuing the upgrade to fiber and other cable companies will all eventually consider it.

There are always skeptics of the need for big bandwidth and many in the industry scoff at gigabit broadband today as nothing more than a marketing ploy. What the critics ignore is that the world grows into larger bandwidth over time. Residential broadband usage is currently growing at a rate of about 21% annually in terms of both total monthly downloads and of desired customer speeds. When gigabit products were first introduced, they were 40 times faster than the average broadband product at that time of about 25 Mbps.

There will inevitably be new uses of bandwidth that will require faster speeds. Just as one example, I saw that Verizon had acquired the products of the augmented reality firm Jaunt. We have all been promised the future ability to hold virtual hologram meetings, and when somebody develops such a product it’s going to sweep the country.  When that happens, households will be bouncing up against the gigabit speed limit and asking for more. I also ask the cable companies to not forget my holodeck – I’m still waiting.

G.Fast over Coax

There is yet another new technology available to carriers – G.Fast over coaxial cable. Early trials of the technology show it works better than G.Fast over telephone copper.

Calix recently did a test of the new coaxial technology and was able to deliver 500+ Mbps for up to 2,000 feet. This is far better than current G.Fast technology over copper which can handle similar data speeds up to about 800 feet. But telephone G.Fast is improving and Calix just demonstrated a telephone copper G.Fast that can deliver 1 Gbps for about 750 feet.

But achieving the kinds of speeds demonstrated by Calix requires a high-quality telephone copper network. We all know that the existing telephone and coaxial networks in existing buildings are usually anything but pristine. Many existing coaxial cables in places like apartment buildings have been cut and re-spliced numerous times over the years, which will significantly degrade G.Fast performance.

This new technology is definitely going to work best in niche applications – and there may be situations where it’s the clearly best technology for the price. There are a surprising number of coaxial networks in place in homes, apartment buildings, schools, factories and older office buildings that might be good candidates for the technology.

A number of telcos like CenturyLink and AT&T are starting to use G.Fast over telephone copper to distribute broadband to apartment buildings. Since as the incumbent telephone company they can make sure that these networks are available to them. But there might be many apartment buildings where the existing coaxial network could be used instead. The ability to go up to 2,000 feet could make a big difference in larger apartment buildings.

Another potential use would be in schools. However, with the expanding demand for broadband in classrooms one has to wonder if 500 Mbps is enough bandwidth to serve and share among a typical string of classrooms – each with their own heavy broadband demand.

There are also a lot of places that have coaxial networks that you might not think about. For example, coaxial wiring was the historic wiring of choice for the early versions of video surveillance cameras in factories and other large businesses. It would not be hard to add WiFi modems to this kind of network. There are tons of older hotels with end-to-end coaxial networks. Any older office buildings is likely to have coaxial wiring throughout.

But there is one drawback for the technology in that the coaxial network can’t be carrying a cable TV signal at the same time. The coaxial G.Fast operates at the same frequencies as a significant chunk of a traditional DOCSIS cable network. To use the technology in a place like an apartment would mean that the coaxial wiring can no longer be used for cable TV delivery. Or it means converting the cable TV signal to IPTV to travel over the G.Fast. (but that wouldn’t leave much bandwidth for broadband.) But still, there are probably many unused coaxial wiring networks and the technology could use them with very little required rewiring.

It’s more likely that the coaxial G.Fast could coexist with existing applications in places like factories. Those networks typically use MoCA to feed the video cameras, at frequencies that are higher than DOCSIS cable networks.

But my guess is that the interference issue will be a big one for many potential applications. Most apartments and schools are going to still be using their networks to deliver traditional video. And many other coaxial networks will have been so chopped up and re-spliced over time to present a real challenge for the technology.

But this is one more technology to put into the toolbox, particularly for companies that bring broadband to a lot of older buildings. There are probably many cases where this could be the most cost effective solution.

Primer on DOCSIS

OLYMPUS DIGITAL CAMERAAnybody who uses a cable modem at home has probably heard the word DOCSIS. This is a set of standards that define how data is transmitted over coaxial cable networks. DOCSIS stands for Data Over Cable Service Interface Specification. It was developed by CableLabs, which is a research and standards organization that the cable companies have created for research and development purposes. CableLabs is to the large cable TV companies what Bell Labs has been for the large telephone companies.

DOCSIS 1.0 was first issued in 1997 as a standard and created the basis for cable modems. It established a data network that started with a CMTS (cable modem terminal system) that talked to cable modems in each home. DOCSIS 1.0 was limited to a single data channel which means that data speeds were limited to a usable 38 Mbps download and 9 Mbps upload for everybody together on a cable node. Because the data was shared with anywhere up to 200 homes, speeds on DOCSIS 1.0 were practically limited to a maximum of about 7 Mbps, although these speeds could be much slower at peak times.

The standard was updated in 1999 to DOCSIS 1.1 which allowed for QoS (Quality of Service) which enabled cable systems to carry voice calls, with priority, on the cable modem data path. There are still a significant number of field deployments using DOCSIS 1.0 and 1.1, particularly in smaller and rural cable systems.

DOCSIS 2.0 came out in 2001 and the major improvement was to increase upload speeds. Version 2.0 also improved the ability to transmit VoIP. The standard still kept the single channel downstream. As cable companies lowered node sizes there were DOCSIS 2.0 systems that supported speeds of up to 15 Mbps download.

The biggest improvement with DOCSIS came with version 3.0 which was released in August 2006. This standard allows for bonding cable channels together to make larger data paths to each node. Cable companies that have deployed DOCSIS 3.0 are offering much faster speeds than in the past. Comcast in the US offers 107 Mbps download in urban markets using the newer modems. In Canada, Shaw Cable and Videotron have used DOCSIS 3.0 to offer products over 200 Mbps download. Virgin Media in Britain announced a speed of 1.5 Gbps download and 150 Mbps upload.

Why don’t US cable companies offer speeds that fast? There is a trade-off in any cable system between the number of channels that are used for programming versus data. While US cable companies have undergone digital conversion to free up channels, they have used most of that new space to add high definition channels to their network rather than dedicate the extra space to data. In the future, cable companies will be able to free up even more space for data by converting their cable channels to IPTV. Today they multicast every channel in the system to every home, but with IPTV they would send only the channels people are watching.

The CEOs of the largest cable companies have often been quoted saying that they are providing the bandwidth that people need. And I am sure that the believe this. But we have a very long way to go to just convert all of the cable systems in the US to DOCSIS 3.0 and increase speeds. I work every day in markets where the speeds are far slower than they are in upgraded urban markets. But it’s good to know that the tools are there for the cable systems to increase speeds, when they finally decide the time is right to do so.

We Don’t Have Enough Bandwidth

I read three different articles Friday that have a common theme – we just don’t have enough bandwidth in this country.

The first article from the Fiber To The Home Council which reports on a recent survey. They report that video viewing over the Internet is growing faster than expected, led by the viewing habits of the young. One third of young viewers watch video on a cell phone or tablet at the same time that they watch TV. And 12% of viewers under 35 report watching all of their content over the Internet.

The article also points out a recent report from Conviva, a web optimization company, who reports that they sampled 22.6 billion video streams and found that 60% of them suffered some degradation due to inadequate bandwidth.

The gist of the article is that demand keeps growing while many parts of the Web are near or at a breaking point in terms of capacity and quality. It’s also evidence that homes don’t want to just watch streaming video, they want to watch multiple streaming videos.

In another article Time Warner announced that it would roll out significantly faster Internet service, but only in competitive markets. The upgrades will come in markets where they are competing against fast competition, such as places where Verizon has built FiOS, where AT&T has relatively fast U-verse and where municipalities have built fiber networks. The company says that they will upgrade to DOCSIS 3 and also install much faster wireless routers. They also will upgrade the DVRs in these markets and roll out apps that are designed for the faster Internet.

But Time Warner also made it clear that they have no plans to upgrade markets where there is not fast competition. My take away from this article is that a lot of the incumbent providers are still only doing upgrades in response to direct competition. Otherwise they are quite satisfied with the status quo and only make investments under duress.

Finally, the citizens of Bergen County, New Jersey have started a petition to ask their politicians to offer whatever is necessary to attract Google fiber to the county. Bergen County is the most populous county in the state.

I find this somewhat surprising because most of the people in this county have Verizon FiOS available. And recently Verizon said they plan to have all of New Jersey covered by FiOS. Most of the rest of the country would be thrilled to be upgraded to the kinds of speeds available in Bergen County. FiOS speeds differ by market, but most markets have speeds available from 15 Mbps download to 150 Mbps download. And a few markets have 300 Mbps and 500 Mbps speeds available. Of course, Google would be bring 1 Gbps speeds for a little more than what people are paying for 50 Mbps from Verizon.

My takeaway from this is that people are beginning to realize how important very fast Internet service is. Even those who already have some of the fastest Internet speeds in the country do not view what they have as a value.

Unfortunately for the citizens of Bergen County I find it highly unlikely that Google will ever build to compete against another fiber network. Verizon could easily upgrade their network to compete with Google on speed and price and the conventional wisdom is that nobody is going to build a second fiber network to homes or both fiber owners will go broke competing against each other.

But all of these articles are indicative of the daily articles I see that continue to highlight the big gap between the bandwidth people want and what they are being offered in the market. We just don’t have enough bandwidth in the country, at least according to consumers.

How Much Bandwidth Can a Cable TV System Deliver?

Cut showing the composition of a coaxial cable.

Cut showing the composition of a coaxial cable. (Photo credit: Wikipedia)

There are a number of techniques that are available for a traditional cable TV network to upgrade the bandwidth on the network available for customer data. If you are operating or competing against a cable TV system you should recognize that there are a number of upgrades that can when combined can drastically improve data speeds. Each of these upgrades comes at a cost, but you can’t discount the technical capabilities of an HFC network if data delivery becomes the primary goal of the network.

  1. Increase System Bandwidth. An example of this kind of upgrade is when a system is upgraded from 750 MHz to 1,000 MHz (or 1 GHz). This upgrade provides more bandwidth by widening the frequencies that are available on the coax. A system bandwidth can be a major upgrade and can involve replacing all of the power taps in the system, and in some systems even requires replacing the coaxial cable.
  1. Reducing Node Size. A node in an HFC system is a neighborhood of homes and/or businesses that share the same bandwidth. Typically there is fiber built to a node and then coax cable from the node to each customer. Historically, before cable modems, nodes were large, often at 1,000 homes or more. But many cable companies have deployed more fiber and reduced node sizes and some cable companies now have nodes in the 200 customer range. Making smaller nodes creates smaller pools of shared bandwidth, meaning there is more bandwidth available to customers at peak times.
  1. MPEG4 Compression. A lot of cable systems still use a compression technique known as MPEG2. This technology is used to compress the digital channels on a network today so that up to ten digital channels will fit into one 6 MHz analog slot. But with MPEG4 as many as 20 digital channels can fit into the same 6 MHz slot. The biggest issue with this conversion is that older set-top boxes won’t recognize MPEG4.
  1. Deploy DOCSIS 3.0. DOCSIS 3.0 is a bandwidth management technology that allows a cable modem to use a larger window of RF frequency for data. The way this works is that a cable system can ‘bond’ multiple channel slots together to that the cable modems can use more than one 6 MHz channel slow for data.
  1. Migrate Analog Channels to Digital. A cable provider can gain some bandwidth space by migrating analog channels to an existing digital line-up. There are often contractual requirements with programmers that make this difficult to achieve. However, as mentioned above, as many as 20 digital channels can fit in the same sized slot as an analog channel. There are always customer issues to also consider since this kind of conversion will shrink the analog offering and expand the digital tiers.
  1. Full Digital Conversion. In a full digital conversion all channels are converted to digital. Once completed, every customer needs a set-top box or other device in order to decode and view channels. There is now a device called a Digital Television Adapter (DTA) that is less costly than a set-top box that can support a customer remote. It is possible to send the ‘basic’ channels through the network un-encoded so that customers with a digital QAM tuner in their TV will be able to see these channels without a DTA.
  1. Deploy Data QOS. This technique does not increase system bandwidth, but rather allows the cable provider to sell faster data to some customers by allowing those customers to use a frequency allocation that is only used by these faster data customers. For example, Comcast advertises 100 Mbps service in most large cities, and they would deliver that kind of speed by giving the 100 Mbps customer priority over other customers in the node by having those customers send their data over a lesser-used frequency on the COAX. Of course, as the priority customer gets more bandwidth, everybody else in the node gets degraded service, and if too many premium services are sold then even the priority customer can’t get the promised bandwidth. But this technique does allow the cable company to selectively compete against fiber for selected customers willing to pay for the extra speed.
  1. Convert to IPTV. This conversion would allow a cable system to use more of the RF frequency on the network for bandwidth. On an IPTV system the programming, voice and cable modem service are all sent over shared bandwidth. An IPTV conversion does not automatically gain a lot of extra bandwidth and any savings come from the fact that the company does not have to broadcast all channels to all nodes all of the time, but rather can just those channels that somebody in the node is watching. There is a benefit, but it is not as large as the extra bandwidth gained by other strategies.
  1. Higher Spectral Efficiency. This technique involves converting to DOCSIS 3.1 and also changing the system modulation techniques. The traditional modulation technique is called QAM (Quadature Amplitude Modulation) and uses a 6 MHz frequency allocation.  The new technique is ODFM (Orthogonal Frequency Division Multiplexing) which uses a higher QAM modulation.  Where Current DOCSIS capabilities achieve approximately 6.3 bits per Hertz, DOCSIS 3.1 can achieve 10 bits per Hertz. New modulation techniques can create much larger bandwidth slots and can at the same time increase the bits to Hz efficiency of the frequency being used. In effect, this technology turns the cable system into a DSL system, with the difference being that there is more frequency available on a coaxial cable than is available on a telephone copper cable, but that a CATV node is then shared by multiple subscribers.

As can be seen, a cable company has a lot of options to increase bandwidth. So, how much bandwidth can be delivered? There are a lot of cable networks that have been upgraded through step 7 above. These systems can support some selected customers up to 100 Mbps download. But these systems probably only support 30 Mbps for all subscribers if the nodes are small enough. A system that is upgraded through step 8 can probably deliver 50 – 60 Mbps to most customers with selected customers being able to get much faster speeds. But a full upgrade to through step nine would allow a cable system to match the overall bandwidth delivered by a fiber PON system, although it is then shared with a lot more customers.

These upgrades are expensive. But if you are competing against a cable company, don’t assume that they are incapable of delivering very decent internet speeds if they are willing to make enough investment in their network.

If you have questions or want to discuss this further call Derrel Duplechin at CCG at (337) 654-7490.