Zen

Translation

The `congestion collapse` is a problem that all heterogeneous networks face. Different mechanisms have been proposed in the scientific literature to avoid or control network congestion. Some of them have been implemented and deployed in real networks. To understand this problem in more detail, let us first consider a simple network with two hosts attached to a high bandwidth link that are sending segments to destination `C` attached to a low bandwidth link as depicted below.
0/4790 · 479

Sign in to save translations.

English Czech Actions
802.5 data frame format
 
To capture a token, a station must operate in `Listen` mode. In this mode, the station receives bits from its upstream neighbor. If the bits correspond to a data frame, they must be forwarded to the downstream neighbor. If they correspond to a token, the station can capture it and transmit its data frame. Both the data frame and the token are encoded as a bit string beginning with the `Starting Delimiter` followed by the `Access Control` field. When the station receives the first bit of a `Starting Delimiter`, it cannot know whether this is a data frame or a token and must forward the entire delimiter to its downstream neighbor. It is only when it receives the fourth bit of the `Access Control` field (i.e. the `Token` bit) that the station knows whether the frame is a data frame or a token. If the `Token` bit is reset, it indicates a data frame and the remaining bits of the data frame must be forwarded to the downstream station. Otherwise (`Token` bit is set), this is a token and the station can capture it by resetting the bit that is currently in its buffer. Thanks to this modification, the beginning of the token is now the beginning of a data frame and the station can switch to `Transmit` mode and send its data frame starting at the fifth bit of the `Access Control` field. Thus, the one-bit delay introduced by each Token Ring station plays a key role in enabling the stations to efficiently capture the token.
 
After having transmitted its data frame, the station must remain in `Transmit` mode until it has received the last bit of its own data frame. This ensures that the bits sent by a station do not remain in the network forever. A data frame sent by a station in a Token Ring network passes in front of all stations attached to the network. Each station can detect the data frame and analyze the destination address to possibly capture the frame.
 
The text above describes the basic operation of a Token Ring network when all stations work correctly. Unfortunately, a real Token Ring network must be able to handle various types of anomalies and this increases the complexity of Token Ring stations. We briefly list the problems and outline their solutions below. A detailed description of the operation of Token Ring stations may be found in [IEEE802.5]_. The first problem is when all the stations attached to the network start. One of them must bootstrap the network by sending the first token. For this, all stations implement a distributed election mechanism that is used to select the `Monitor`. Any station can become a `Monitor`. The `Monitor` manages the Token Ring network and ensures that it operates correctly. Its first role is to introduce a delay of 24 bit transmission times to ensure that the token can travel smoothly on the ring. Second, the `Monitor` sends the first token on the ring. It must also verify that the token passes regularly. According to the Token Ring standard [IEEE802.5]_, a station cannot retain the token to transmit data frames for a duration longer than the `Token Holding Time` (THT) (slightly less than 10 milliseconds). On a network containing `N` stations, the `Monitor` must receive the token at least every :math:`N \times THT` seconds. If the `Monitor` does not receive a token during such a period, it cuts the ring for some time and then re-initializes the ring and sends a token.
 
Several other anomalies may occur in a Token Ring network. For example, a station could capture a token and be powered off before having resent the token. Another station could have captured the token, sent its data frame and be powered off before receiving all of its data frame. In this case, the bit string corresponding to the end of a frame would remain in the ring without being removed by its sender. Several techniques are defined in [IEEE802.5]_ to allow the `Monitor` to handle all these problems. If unfortunately, the `Monitor` fails, another station will be elected to become the new `Monitor`.
 
Congestion control
 
Most networks contain links having different bandwidth. Some hosts can use low bandwidth wireless networks. Some servers are attached via 10 Gbps interfaces and inter-router links may vary from a few tens of kilobits per second up to hundred Gbps. Despite these huge differences in performance, any host should be able to efficiently exchange segments with a high-end server.
 
To understand this problem better, let us consider the scenario shown in the figure below, where a server (`A`) attached to a `10 Mbps` link needs to reliably transfer segments to another computer (`C`) through a path that contains a `2 Mbps` link.
 
In this network, the segments sent by the server reach router `R1`. `R1` forwards the segments towards router `R2`. Router `R1` can potentially receive segments at `10 Mbps`, but it can only forward them at `2 Mbps` to router `R2` and then to host `C`. Router `R1` includes buffers that allow it to store the packets that cannot immediately be forwarded to their destination. To understand the operation of a reliable transport protocol in this environment, let us consider a simplified model of this network where host `A` is attached to a `10 Mbps` link to a queue that represents the buffers of router `R1`. This queue is emptied at a rate of `2 Mbps`.
 
Self clocking
 
However, transport protocols are not only used in this environment. In the global Internet, a large number of hosts send segments to a large number of receivers. For example, let us consider the network depicted below which is similar to the one discussed in [Jacobson1988]_ and :rfc:`896`. In this network, we assume that the buffers of the router are infinite to ensure that no packet is lost.
 
If many senders are attached to the left part of the network above, they all send a window full of segments. These segments are stored in the buffers of the router before being transmitted towards their destination. If there are many senders on the left part of the network, the occupancy of the buffers quickly grows. A consequence of the buffer occupancy is that the round-trip-time, measured by the transport protocol, between the sender and the receiver increases. Consider a network where 10,000 bits segments are sent. When the buffer is empty, such a segment requires 1 millisecond to be transmitted on the `10 Mbps` link and 5 milliseconds to be the transmitted on the `2 Mbps` link. Thus, the measured round-trip-time measured is roughly 6 milliseconds if we ignore the propagation delay on the links. If the buffer contains 100 segments, the round-trip-time becomes :math:`1+100 \times 5+ 5` milliseconds as new segments are only transmitted on the `2 Mbps` link once all previous segments have been transmitted. Unfortunately, if the reliable transport protocol uses a retransmission timer and performs `go-back-n` to recover from transmission errors it will retransmit a full window of segments. This increases the occupancy of the buffer and the delay through the buffer... Furthermore, the buffer may store and send on the low bandwidth links several retransmissions of the same segment. This problem is called `congestion collapse`. It occurred several times during the late 1980s on the Internet [Jacobson1988]_.
 
The `congestion collapse` is a problem that all heterogeneous networks face. Different mechanisms have been proposed in the scientific literature to avoid or control network congestion. Some of them have been implemented and deployed in real networks. To understand this problem in more detail, let us first consider a simple network with two hosts attached to a high bandwidth link that are sending segments to destination `C` attached to a low bandwidth link as depicted below.
 
To avoid `congestion collapse`, the hosts must regulate their transmission rate [#fcredit]_ by using a `congestion control` mechanism. Such a mechanism can be implemented in the transport layer or in the network layer. In TCP/IP networks, it is implemented in the transport layer, but other technologies such as `Asynchronous Transfer Mode (ATM)` or `Frame Relay` include congestion control mechanisms in lower layers.
 
Let us first consider the simple problem of a set of :math:`i` hosts that share a single bottleneck link as shown in the example above. In this network, the congestion control scheme must achieve the following objectives [CJ1989]_ :
 
The congestion control scheme must `avoid congestion`. In practice, this means that the bottleneck link cannot be overloaded. If :math:`r_i(t)` is the transmission rate allocated to host :math:`i` at time :math:`t` and :math:`R` the bandwidth of the bottleneck link, then the congestion control scheme should ensure that, on average, :math:`\forall{t} \sum{r_i(t)} \le R`.
 
The congestion control scheme must be `efficient`. The bottleneck link is usually both a shared and an expensive resource. Usually, bottleneck links are wide area links that are much more expensive to upgrade than the local area networks. The congestion control scheme should ensure that such links are efficiently used. Mathematically, the control scheme should ensure that :math:`\forall{t} \sum{r_i(t)} \approx R`.
 
The congestion control scheme should be `fair`. Most congestion schemes aim at achieving `max-min fairness`. An allocation of transmission rates to sources is said to be `max-min fair` if :
 
no link in the network is congested
 
the rate allocated to source :math:`j` cannot be increased without decreasing the rate allocated to a source :math:`i` whose allocation is smaller than the rate allocated to source :math:`j` [Leboudec2008]_ .
 
Depending on the network, a `max-min fair allocation` may not always exist. In practice, `max-min fairness` is an ideal objective that cannot necessarily be achieved. When there is a single bottleneck link as in the example above, `max-min fairness` implies that each source should be allocated the same transmission rate.
 
To visualize the different rate allocations, it is useful to consider the graph shown below. In this graph, we plot on the `x-axis` (resp. `y-axis`) the rate allocated to host `B` (resp. `A`). A point in the graph :math:`(r_B,r_A)` corresponds to a possible allocation of the transmission rates. Since there is a `2 Mbps` bottleneck link in this network, the graph can be divided into two regions. The lower left part of the graph contains all allocations :math:`(r_B,r_A)` such that the bottleneck link is not congested (:math:`r_A+r_B<2`). The right border of this region is the `efficiency line`, i.e. the set of allocations that completely utilize the bottleneck link (:math:`r_A+r_B=2`). Finally, the `fairness line` is the set of fair allocations.
 
Possible allocated transmission rates
 
As shown in the graph above, a rate allocation may be fair but not efficient (e.g. :math:`r_A=0.7,r_B=0.7`), fair and efficient ( e.g. :math:`r_A=1,r_B=1`) or efficient but not fair (e.g. :math:`r_A=1.5,r_B=0.5`). Ideally, the allocation should be both fair and efficient. Unfortunately, maintaining such an allocation with fluctuations in the number of flows that use the network is a challenging problem. Furthermore, there might be several thousands flows that pass through the same link [#fflowslink]_.
 
To deal with these fluctuations in demand, which result in fluctuations in the available bandwidth, computer networks use a congestion control scheme. This congestion control scheme should achieve the three objectives listed above. Some congestion control schemes rely on a close cooperation between the end hosts and the routers, while others are mainly implemented on the end hosts with limited support from the routers.
 
A congestion control scheme can be modeled as an algorithm that adapts the transmission rate (:math:`r_i(t)`) of host :math:`i` based on the feedback received from the network. Different types of feedback are possible. The simplest scheme is a binary feedback [CJ1989]_ [Jacobson1988]_ where the hosts simply learn whether the network is congested or not. Some congestion control schemes allow the network to regularly send an allocated transmission rate in Mbps to each host [BF1995]_.
 
Let us focus on the binary feedback scheme which is the most widely used today. Intuitively, the congestion control scheme should decrease the transmission rate of a host when congestion has been detected in the network, in order to avoid congestion collapse. Furthermore, the hosts should increase their transmission rate when the network is not congested. Otherwise, the hosts would not be able to efficiently utilize the network. The rate allocated to each host fluctuates with time, depending on the feedback received from the network. The figure below illustrates the evolution of the transmission rates allocated to two hosts in our simple network. Initially, two hosts have a low allocation, but this is not efficient. The allocations increase until the network becomes congested. At this point, the hosts decrease their transmission rate to avoid congestion collapse. If the congestion control scheme works well, after some time the allocations should become both fair and efficient.
 

Loading…

User avatar None

Source string added
The `congestion collapse` is a problem that all heterogeneous networks face. Different mechanisms have been proposed in the scientific literature to avoid or control network congestion. Some of them have been implemented and deployed in real networks. To understand this problem in more detail, let us first consider a simple network with two hosts attached to a high bandwidth link that are sending segments to destination `C` attached to a low bandwidth link as depicted below.
6 years ago
Browse all string changes

Glossary

English Czech
No related strings found in the glossary.

18607 String information

Source string location
../../principles/sharing.rst:727
String age
6 years ago
Last updated
2 months ago
Source string age
6 years ago
Translation file
locale/cs/LC_MESSAGES/principles/sharing.po, string 154