Next: THE ACTS CONNECTION
Up: NETWORK ARCHITECTURE
Previous: Network protocol layer
ATM is not a ``reliable'' protocol in the networking sense - cells may
be dropped or lost in transit, and no attempt is made to verify
delivery. In order to facilitate reliable data transfer, and to allow
the use of the wealth of software tools already available, we are
running the standard IP protocols over ATM. We are using a
pseudo-standard implementation known as ``Classical IP'', which
defines a relationship between standard IP ``dot'' addressing and ATM
PVCs, as well as data packet segmentation and reassembly algorithms
for converting between IP packets and ATM cells[6].
Since the ATM switches know nothing of the protocol levels above ATM,
the choice of this protocol affects merely the two endpoint systems.
Those workstations both employ FORE SBA-200 ATM interface cards which
perform IP packet segmentation and reassembly in hardware, enabling
net ATM speeds of 155 Mbit/sec (OC-3).
Over the Classical IP level, we run the same TCP/IP and UDP/IP
protocols that are used over standard Ethernet networks. Tools such as
ftp, telnet, and the X Window
System
are part of every observing run, as are
additional tools, such as an audio conferencing tool (rat) and a
shared whiteboard application (wb).
The most important impact of a satellite component on a high-speed
network is the relatively large delay introduced by the round-trip
signal travel time to the satellite. In our network, this travel time
is approximately 0.55 seconds, which corresponds to over 3 Mbytes of
data at OC-1 speeds (51 Mbit/sec). The problem has to do with the
connection-oriented nature of TCP/IP: TCP sends a very specific amount
of data, known as a ``window'', after which time it expects an
acknowledgement from the other end of the connection. This is the
manner in which TCP/IP is able to implement a ``reliable'' connection.
However, this window size is often very small; the default value for
workstations running the SunOS 4.1.4 operating system is 4 Kbytes. If
one were to use this system on our satellite network in its default
configuration, a window of data would be sent in 0.0007 seconds, after
which the sending system would be forced to wait 0.549 seconds for an
acknowledgement. In other words, the system would be running at 0.1%
efficiency, and the net throughput would reflect this: initial tests
of our system under such conditions showed bandwidths of
0.1-0.2 Mbit/sec.
Fortunately, this problem is well-known in the high-speed networking
community. Networks such as ours are known as ``long fat networks''
(LFN). The figure of merit for such networks is the window size, or
the bandwidth-delay product:
| |
(1) |
There is an Internet RFC (Request For
Comment)
document on
this subject, RFC 1323[7], and the problem has been
discussed extensively. Many current operating systems support the
RFC 1323 extensions, and provide options to increase the TCP window
size to values in excess of 10 Mbytes. In the case of the SunOS
operating system (to which we are constrained by legacy control
software at Keck), we obtained the TCP-LFN package from Sun
Consulting, which also purports to support the RFC 1323 extensions for
long fat networks. In reality, a number of limitations of SunOS
conspire to prohibit one from obtaining extremely large window sizes.
In our case, the kernel limit of 2 Mbytes of Mbuf memory (IP packet
wrappers) turned out to be the major constraint, limiting our window
size to no more than 1 Mbyte. This is approximately one-third of the
optimal value derived above, and, as such, we obtain maximum TCP/IP
throughputs of approximately 10-15 Mbit/sec, about one-third of the
51 Mbit/sec OC-1 network operating speed. Although perhaps
disappointing in a relative sense, this bandwidth is far in excess of
T1 Ethernet speed (1.44 Mbit/sec) and allows an 8 Mbyte image to be
transferred in approximately 5 seconds (not including protocol
overhead). As a further comparison, this bandwidth exceeds that which
is available on the local area Ethernet network at the Keck Telescope
itself.
Next: THE ACTS CONNECTION
Up: NETWORK ARCHITECTURE
Previous: Network protocol layer
Patrick Shopbell
8/11/1997