atcomsystems.ca/forum Forums Wiring - Telecom Carriers - Trunks & Services T1's, CSU's and DSU's T1 Data 1.54 upload/download at same time or total 1.54?

Hi.

If I have a Data T1, can I upload and download 1.54 mpbs at the same time or Total?

For example, I am uploading a file a 1.5 Mpbs, will I be unable to download another file at 1.54 Mpbs?

or will it be split 768/768?

Thanks.

No, that's the maximum througput of a T1, 24 channels at 64K each. It probably would not be split down the middle, it would probably bounce back and forth depending upon what else is going on at the moment. Sorry.

Oh no! Well, I guess you are right.

So how come Comcast has 6.0 Mbps download and 768Kbps upload? So if you are uploading at 768K, you can't download anything or I guess they are working off a different technology.

If a T1 is nothing more then 2 pair of twisted coper wire, why does it cost so much more than a phone line? Is it more expensive to run the lines?

Anyway, Thanks for your help.

Its not your fault, I should of known better.

a standard T1 provides 24 available channels. These can be used as either plain old lines r trunks, so on your set of wires you have the ability to run 24 seperate lines(with a channel bank) Also downtime is cut down, most carriers offer a 24 to 48 hour repair time for pots lines, and somewhere around a 4 hour turn around for T1 ckts.

Comcast, I assume is either cable or DSL technology. Those technologies are more closely related to ethernet than T1.

A voice T1 is 24 channels of digitized voice conversations. A straight T1 steals a bit here and there from the voice channels. A PRI uses one channel as data, and the 23 others are the same old digitized voice data.

A data-only T1 would be more akin to 24 56k modems tied-together at once. Each channel in a T1 is only realistically capable of 56k-bits per second of throughput, one direction at a time. 24 x 56k is 1.344M-bits. The equipment that interfaces the T1 to the data network will dynamically allocate the channels for each data stream up or down. If there was only one data stream, it would have all available bandwidth. If there were two, they'd each have half, and so on.

But the real trick is that internet traffic is not a one-way street. Both ends of the T1 will group packet data together and send one long burst at higher speed rather than a bunch of short packets each at a slower speed. So the aggregate speed of any one direction is still quite high.

T1's cost so much because of the equipment needed and therefore maintained at each end to support it. For voice, it would typically be cheaper to operate a T1, rather than 24 individula POTS lines. But for data, there are much cheaper technologies available today. DSL, (though limited in availability) cable, and fiber.

The reason that T1 circuits cost so much more is due the fact that they are technically 24 lines, or at least they have that capability. Then, like Anthony said, the maintenance cost is higher for the providers because of the expensive electronics, fast turnaround time for repairs, and generally more intelligent support staff.

The bandwidth on a Comcast is much higher because they are using a completely different technology. Broadband uses the higer capacity of fiber optics and coax to attain this capability.

A data T1 is not normally broken into channels… Called High Caps or a little more obscure term that we use sometimes use is an Incantonated T1. These have become extremely common for business applications.

These are full 1.544 Mb/s digital pipes that are timed and are available to run at all times at that speed and ONLY at that speed. Think of ‘em like a conveyor belt that is always on and ALWAYS moving at that exact speed. It does not matter if you trying to move data on it or not, it’s still there moving at 1.544 Mb/s, ready to take data into or out of your equipment at that rate, all the time.

Let’s say you send 5 ICMP packets to a device on the other end of a High-Cap (send a ping) those 5 packets are whisked away and then hopefully returned with a successful ping test over a 1.544 Mb/s highway.

Now, you want to set up a point –to- point video conference (typically BIG bandwidth chewer-upper)… The digitized image and sound from one end is moved to other at the rate of 1.544 Mb/s. It does not matter that you could actually use more in this application…. 1.544 Mb/s is what you get all the time.

Now what you’re referring to with Comcast “6.0 Mbps download and 768Kbps upload” that’s ADSL… Like T1 these are digital services and are always on… BUT ADSL is NOT a timed circuit and it’s not always moving. In the conveyor belt analogy think of it as a conveyor belt that sits stopped when not being used but nearly instantansly as something is placed on it, it’ll take off. The advertised speeds are “best case” and not set in stone like T1. So again with the conveyor belt is moves at variable speeds dependant on certain factors that are usually near 6.0 Mbps in and 768Kbps out. One more thing to know, the “A” part in ADSL is asynchronies… meaning upload and download speeds do not match.

The installation and maintenance of these data T1’s are WAY more expensive than ADSL; and the biggest cost difference for data T1 over ADSL is the interoffice transport. T1’s have to be MUX-ed and moved from place to place as separate pipes over DS3 time slots to ensure that T1 has a full 1.544 Mb/s available to it all day, everyday. Whereas ADSL circuits hit IMA (inverse multiplexed) pathways… and they can share interoffice bandwidth based on % usage estimates.

Oooopps… Both Ed and Rob responded in the time it took me to type that response all out.

Rob,
I’m going to disagree with this statement… If running T1 AMI it would in fact be like 24 56 Kb modems tied together… When getting a full T1 with B8ZS line coding (as most are T1’s are ordered) the capability is there for it to act like 24 64 Kb modes with the remaining 8 bits and some in-band tricks used for timing and error detection.

I didn't realize data T's weren't AMI. I just remember reading the explanation of why 56k modems were 56 not 64 having to do with the T1 signaling....

No matter... we all had our own level of understanding and input, and in the end an answer to the question arose....

All T1's are AMI (Alternate Mark Inversion) even B8ZS T's. Each pulse in a T1 circuit is of opposite polarity, hence the term alternate mark. This is the most basic form of error indication and is how “polarity violations” are generated. The term AMI is generally accepted to describe a standard line code for non-B8ZS T's. The standard line code is actually zero-code suppression, sometimes called jam-bit 7 or bit stuffing. This coding adds a mark to the 7th bit (also called the least significant bit) in a byte when the byte is all zeros. This is why 56k circuits exist, your through-put is limited. If you would like to use 64k or "clear channel", you need to use B8ZS line coding. This creates a polarity violation on bits 4,5 and 6,7 (if I remember the bits correctly) when all zeros are transmited to keep the 12.5% one-density rule in line. One density is critical to a T1. If more than 15 consecutive zeros are produced, repeaters start to lose timing on the circuit. Line coding keeps the T1 happy and running in sync.