Features And Benefits; Frame Buffering And Latency; Additional Features And Benefits - GE MultiLink ML600 Instruction Manual

CHAPTER 2: PRODUCT DESCRIPTION
2.2

Features and Benefits

2.2.1

Frame Buffering and Latency

2.2.2

Additional Features and Benefits

MULTILINK ML600 ETHERNET COMMUNICATIONS SWITCH – INSTRUCTION MANUAL
The ML600 is store-and-forward switch. Each frame (or packet) is loaded into the switch
memory and inspected before forwarding can occur. This technique ensures that all
forwarded frames are of a valid length and have the correct CRC (i.e., they are good
packets). This eliminates propagation of bad packets, enabling all of the available
bandwidth to be used for valid information.
While other switching technologies such as "cut-through" or "express" impose minimal
frame latency, they also permit bad frames to propagate to the Ethernet network. The
"cut-through" technique permits collision fragment frames, which are a result of late
collisions, to be forwarded – which add to the network traffic. There is no way to filter
frames with a bad CRC (the entire frame must be present in order for CRC to be calculated).
Since collisions and bad packets are more likely when traffic is heavy, store-and-forward
switch technology enables more bandwidth to be available for good packets when the
traffic load is greatest.
To minimize the possibility of dropping frames on congested ports, each ML600
dynamically allocates buffer space from a 128 KB memory pool, ensuring that heavily used
ports receive very large buffer space for packet storage (many other switches have their
packet buffer storage space divided evenly across all ports, resulting in a small, fixed
number of packets to be stored per port; when the port buffer fills up, dropped packets
result). This dynamic buffer allocation provides the capability for the maximum resources
of the ML600 to be applied to all traffic loads, even when the traffic activity is unbalanced
across the ports. Since the traffic on an operating network is constantly varying in packet
density per port and in aggregate density, the ML600 switches are constantly adapting
internally to provide maximum network performance with the least dropped packets.
When the ML600 detects that its free buffer queue space is low, it sends industry standard
(full-duplex only) PAUSE packets out to the devices sending packets to cause "flow control".
This tells the sending devices to temporarily stop sending traffic, which allows a traffic
catch-up to occur without dropping packets. Then, normal packet buffering and
processing resumes. This flow-control sequence occurs in a small fraction of a second and
is transparent to an observer.
Another feature implemented in the ML600 is a collision-based flow-control mechanism
(when operating at half-duplex only). When the ML600 detects that its free buffer queue
space is low, it prevents more frames from entering by forcing a collision signal on all
receiving half-duplex ports in order to stop incoming traffic.
The latency (the time the frame spends in the switch before it is sent along or forwarded to
its destination) of the ML600 varies with the port-speed types. The length of the frame is
variable as it is with all store-and-forward switches. For 10 Mb-to-10 Mb, 10 Mb-to-100 Mb
or 100 Mb-to-10 Mb forwarding, the latency is 15 μs plus the packet time of 10 Mb. For
100 Mb-to-100 Mb forwarding, the latency is 5 μs plus the packet time of 100 Mb.
• Full 10 or 100 Mb switching services for high performance Ethernet: The ML600
provides fast Ethernet switching on all ports. They perform high speed filter/forward
operations on the traffic, giving each port segment a full 10 Mb (or 100 Mb) of
bandwidth.
• Reduces network costs and provide an economical solution: The ML600 offers the
ideal solution to efficiently and inexpensively connect a twisted-pair and fiber network
with 10 or 100 Mb and expand in a convenient and economical way.
PRODUCT DESCRIPTION
2–3