|
PureBytes Links
Trading Reference Links
|
Here is something to consider when trying to adapt some of the high speed
serial port boards to DTN or BMI satellite in hopes of minimizing tickloss.
These boards such as the TE920 are designed to accomplish the high speed by
use of hardware or software flow control. The BMI satellite receiver does
not use any flow control and I have been told neither does the 115K DTN box,
i.e. it is like a data hose and has no internal buffering. After a week of
fooling around with configurations on the TE920 on DTN and BMI on different
machines and software, a friend of mine and I are still experiencing
overflows. That means the buffers(even 64 byte buffers) are not emptied
fast enough before new data arrives. The 920 is an ISA board and in my case
it had gross incompatibility with an internal modem in the adjacent slot and
would cause screwy things to happen like windows would freeze and a constant
beeping error occurred. There were no conflicts with irq's or dma or memory
assignments...chuck the modem...with a nonbuffered port the machine is a
single function machine. Yes, it will do many things simultaneously, but at
the loss of ticks. I just don't think that a 64 or even 128 byte buffer
will solve this problem. Thus, the requirement for a better datamanager and
or a truly buffered board or box as mentioned below. I would like to be
proved wrong on this point.
BR
----- Original Message -----
From: "Bob Fulks" <bfulks@xxxxxxxxxxxx>
To: <ALVIS2JS@xxxxxxx>
Cc: <omega-list@xxxxxxxxxx>
Sent: Wednesday, March 15, 2000 4:20 PM
Subject: Re: DTN Satellite / Serial Card Alternatives
> At 1:31 PM -0500 3/14/00, ALVIS2JS@xxxxxxx wrote:
>
> >Am considering Pacific CommWare TurboExpress 920. But DTN rep also
mentioned
> >alternative serial cards from Boca, Siig and particularly Digi
International.
> > Digi (www.digi.com), which apparently has a 60% market share, claims
their
> >twin port Acceleport Xr, is a superior product that minimizes use of
system
> >resources. At $200 to $250 (street price), it is somewhat more
expensive.
> >
> >Does anyone have knowledge of or experience with these alternatives to
the
> >920, particularly the Digi Xr. Any thoughts appreciated.
>
> The following is from two previous posts by me on this topic.
>
> Bob Fulks
>
> -----
>
> I believe the potential benefit of the TurboExpress card in this
application is in the size of its FIFO (first-in-first-out) buffer.
>
> The BMI box (and perhaps others) sends data at a fixed rate.
>
> The server can accept data at some average rate, presumably related to how
much work it has to do, which is related to how many symbols you are
collecting data on, etc.
>
> If the server is mostly not busy (and the computer is not doing anything
else) then it should be sitting idle most of the time waiting for new data
and should be able to grab every tick from the source. But as it gets
busier, then a tick will sometimes occur while the computer is busy doing
other things and the tick would be lost.
>
> The FIFO buffer helps by storing some number of ticks which allows the
server to grab the next one when it has time to process it. If the buffer is
infinitely large, then so long as the average rate at which the server can
process ticks is higher than the average rate the source produces ticks,
then no ticks are lost. They may be delayed but they are not lost. But if
the buffer size is limited, then in certain cases, it will get full and new
ticks will have no place to go and will be lost.
>
> So the bigger buffer in the chip used in the TurboExpress card should
result in fewer ticks getting lost. But even if the card can process data at
some very high speed, the speed of the total process is limited to the
slowest speed of any of the three components, the source, the card, or the
server/computer combination.
>
> To quote the description on their web page at
<http://www.turbocom.com/texport.html>:
>
> Reliability is solid even at extremely high data rates because the
> 16750 UART holds incoming and outgoing characters in extra-deep
> FIFO (First-In-First-Out) buffers. These buffers can store 64 bytes
> each, versus only 16 bytes for the ordinary 16550 UART. UART-mediated
> hardware flow control provides nearly complete immunity from overrun
> and queue overflow errors.
>
> Putting a few numbers on this should give us an idea of how it works.
Assuming a 38.4K baud rate, this is about 4000 bytes per second, or 0.25
millisecond per byte. So the 16 byte FIFO in the 16550 UART chip can store
data for about 4 milliseconds (four thousands of a second). The 64 byte FIFO
in the 16750 UART chip can store data for about 16 milliseconds. So if the
server can process data at greater than 4000 bytes per second on the
average, and doesn't get stuck on some task for more than about 16
milliseconds, the FIFO buffer in the UART chip should allow no data loss.
(Exact calculations would require estimated of the distributions and
queueing theory.)
>
> Assuming we wanted no data loss and could wait as long as a second for the
data, we would ideally like a FIFO 64 times this large, or about 4,000
bytes. No such card is available that I know of. Black Box Corp. has a box
that has a 32,000 byte buffer which is probably overkill for this
requirement, and a lot more expensive.
>
> Bob Fulks
>
> -----------
>
> At 11:11 AM -0600 1/22/00, Terry Wyss wrote:
>
> >There are other manufacturers of serial interface cards that have
> >buffering. I use a Multitech Intelligent Serial Interface model
> >ISI551PC. It has 32k bytes of ram, has physical jumpers to set the
> >IRQ's, is an ISA card, transmits at 115.2K bps, can be accessed by
> >anything that would use a UART 16550. Mine has been reliable and is a
> >couple years old. The ISI processor on the card [to quote the manual]
> >"handles all of the byte-by-byte interrupts generated by the
> >asychronous terminals and stores the data in buffers. The ISI then
> >generates one interrupt for an entireblock of information and
> >transfers the block to the system's microprocessor." There are faster
> >cards with larger buffers, but they are probably overkill.
>
> I found the information at
>
> <http://www.multitech.com/products/FamilyHomePages/tn.asp?ID=24#more>
>
> There appear to be two series of cards, the older 500 series cards
mentioned above and a newer 600 series cards. The older series is ISA/EISA
bus while the newer series has both ISA and PCI versions
>
> 600-Series Features
> Four or eight buffered serial ports
> Sustained data rates to 460.8K per port
> Drivers for Windows 95/98/2000, Windows NT, Citrix, Novell,
> SCO, Linux, and Multi-Tech RASExpress
> PCI- and ISA-bus models available
> 256K RAM buffer
>
> Model Description
> ISI4604-PCI 4-Port Serial Card (PCI)
> ISI4608-PCI 8-Port Serial Card (PCI)
> ISI4608 8-Port Serial Card (ISA)
>
>
> 500-Series Features
> One or two COM ports
> Sustained data rates to 115.2K per port
> Operating-system independent for flexible compatibility
> ISA-bus
> 32K Bytes RAM buffer per port
>
> Model Description
> ISI552 Dual COM-Port Card (ISA)
> ISI551 COM-Port Card (ISA)
>
> A 32K byte buffer would store about 8 seconds worth of BMI 38.4K baud data
which should certainly be sufficient for most purposes.
>
> No prices and no on-line sales that I could see but they will refer you to
dealers. Their phone number is 1-800-328-9717. The office is closed today.
>
> Bob Fulks
>
>
|