Veracity Networks was kind enough to invite me down to their offices to see what they’ve been working on and chew the fat about broadband in Utah. (In the interest of full disclosure, they bought me a tasty but inexpensive lunch from Lon’s Cookin’ Shack. Appreciated, but not enough to buy any influence.) I’ve seen a lot of improvements down there and I feel a lot better about the direction the network is headed in.
You probably recall the Daily Herald running an article a few months ago regarding network upgrades that Veracity had put in place. The specifics are that they have dropped in a lot of new network gear that can handle 10Gbit on the backbone and 1Gbit to individual locations. It’s a nice capacity bump and allows them to better manage the services they have. It also didn’t hurt to upgrade from Layer 2 to Layer 3 routers.
They’ve also had to do a lot of work on getting video up to snuff. The plant down there is still using MPEG-2 which gobbles up 40Mbit per stream, on average. With 100Mbit connections to each home, it’s easy to see how that can disappear very, very quickly. Something that was causing a lot of problems in this regard is that bandwidth caps for the various tiers weren’t functioning properly. It was very easy to saturate the pipe with data and cause some pretty severe video issues. And speaking of video, the customers in Traverse Mountain finally have access to it after many frustrating years.
In all of this, Veracity has put a lot of new monitoring systems in place to catch problems much more quickly. In addition to this, they can also tap directly into each distribution point to see what a customer sees without even leaving the NOC and have setup a full testbed to do trial runs on firmware updates before hitting customers with them. I also got the feeling that a lot of the Broadweave employees who stayed on are being better utilized, having previously been subject to management that didn’t know how to take advantage of an employee’s strengths.
All of the work that Veracity has done has been instead of actively pursuing a lot of new business, and that makes sense. You don’t want to sell some wine before its time because a lot of the potential customers have previously had bad experiences with other service providers. Veracity did say outright that the commercial market in Provo is pretty well saturated at this point, so they’re focusing growth on picking up residential customers. With Qwest’s nearly non-existent network investment (which will no doubt get worse as CenturyLink steps into it) and Comcast’s ho-hum product offerings, I think Veracity’s plan to make it up on volume is a winner. I obviously can’t see the books, but their management team didn’t just fall off the turnip truck. Even if they don’t have the network fully self-sufficient, Veracity will still be hanging onto it. There is plenty of revenue from other operations to cross-subsidize it and having a full fiber infrastructure ahead of the need for it is a smart move.
I’m still kind of sad that Provo didn’t follow the recommendations to make the network self-sufficient. Some relatively minor tweaks would have balanced the books and still kept an open-access network. After Broadweave’s incessant bungling, though, Veracity’s offer is probably the best the city could do. And it’s actually pretty good.
40 Mbits per stream with MPEG2? OTA 1080i MPEG2 broadcasts only go up to 18Mbps. Using MPEG4 or h.264 reduces your bandwidth requirements even further.
See reference here.
40Mbit is what the CTO said it uses. Could be that the older headend equipment isn’t as efficient as it could be. I’m sure it won’t be too long before they make the jump to MPEG-4, but even then a 1080p stream can chew from 16-24Mbit.
40 Mbps is not accurate.
When the city was running the network, it was ~20Mbps for HD streams and ~4Mbps for standard def.
After protocol overhead that 20meg could easily be double. The TCP/IP packet headers alone will had around 15-30% themselves, And who knows what all else they are running to maintain their QoS, monitor feed security, etc etc.
I must say that as someone who knows, 40MB video streams have never happened at iProvo. Mis-communication? The HD streams went as high as 22MB. I have to wonder where some of this nonsense comes from. Of course these 1GB connections must be few and far between since virtually all of their CPEs top out at 100 MB. Let’s be real.
JWilson, I thought I would clear up the HD MPEG2 info:) When Jesse and I went to lunch it was pretty informal and we talked about alot of different things. One of the topics we talked about was why some of the extreme tier residential internet packages were removed.
My answer was that when we sold triple play products to our customer our average customer needed at least two set top boxes and in many cases more. Since our largest MPEG 2 HD streams eat 22 megs per second a user could easily eat 44, to 66 megs per second of the 100 megs we push towards an average residential customer. I think this is where the 40 meg MPEG 2 streams came from. The good news is that we are working with some of our content providers to receve the video streams in both MPEG4 and MPEG2. Soon if a customer has an HD Set top box it will only receive MPEG4 steams and thus negate the problem above.
It is true that most of our links to our residential customer are still 100 meg. But if a customer has need it is very easy for us to push 1 GIG. We have lots of customers that are receiving multi-gig VPLS and PBB connections over the network now.
Hope this clears a few things up.
Thanks!
An additional problem was that the STB is rarely turned off by customers and the boxes can continue to draw the bandwidth of the last channel they were tuned to.
Thus a customer can draw 20 Mb/sec. for each STB tuned to a HDTV channel even when not being watched.
In most homes this would not be a problem as watching 3 full HD channels could be 60 Mb/sec. still leaving more data bandwidth than 95+% of the customers use. (Not all HD channels are 20 Mb/sec. but moving HD to MPEG4 is a good way to conserve bandwidth.)