- Simplicity versus Flexibility versus Optimality
- Knowing the Problem You're Trying to Solve
- Overhead and Scaling
- Operation Above Capacity
- Compact IDs versus Object Identifiers
- Optimizing for the Most Common or Important Case
- Forward Compatibility
- Migration: Routing Algorithms and Addressing
- Making Multiprotocol Operation Possible
- Running over Layer 3 versus Layer 2
- Determinism versus Stability
- Performance for Correctness
- In Closing
There are various reasons for having parameters, some good and others bad.
The protocol designers can not figure out reasonable values, so they leave it to the user. This might make sense if deployment experience might help people determine reasonable values. However, if the protocol designers simply can't decide, it is unreasonable to expect the users to have better judgment. At any rate, if deployment experience gives people enough information to set the values, it would be nice to make the parameters constants in later versions of the protocol.
The values are parameters because there are reasonable trade-offs—say, between responsiveness and overhead—and it is desirable to allow the user to tune the network. In this case, the parameter descriptions should explain the range and the reasons for choosing points in the range.
18.9.1 Avoiding Parameters
In general, it is a good idea to avoid parameters whenever possible because it makes for intimidating documentation that must be written—and, what's more, read—to use the protocol. Sometimes a parameter is an unnecessary frill, and it is possible to pick a value and let everyone live with it. In other cases the parameter can be figured out by the computer rather than the human, and protocols should try hard to make everything plug and play. For example, it is highly desirable for addresses to autoconfigure. Admittedly, autoconfiguration is easier with a protocol such as IPX, but it was well worth the complexity and expense to deploy DHCP to make IP largely autoconfigure.
Another parameter that can autoconfigure is link cost. IPXWAN is a protocol that measures the round trip delay and bandwidth when a link starts up and, from that, calculates a reasonable link cost.
18.9.2 Legal Parameter Setting
It is important to design a protocol so that parameters set by people can be modified in a running network, one node at a time.
In some protocols, parameters can be set incorrectly and the protocol will not run properly. Unfortunately, it isn't as simple as having a legal range for the parameter because one parameter might interact with another, even a parameter in a different layer. In a distributed system it's possible for two systems to independently have reasonable parameter settings but have the parameter settings be incompatible. A simple example of incompatible settings is in a neighbor aliveness detection protocol, in which one sends hellos every n seconds and the other declares the neighbor dead if it does not hear a hello for k seconds. If k is not greater than n, the protocol will not work very well.
There are some tricks for causing parameters to be compatible in a distributed system. In some cases, it is reasonable for nodes to operate with different parameter settings, as long as all the nodes know the parameter setting of other (relevant) nodes.
18.104.22.168 "Report My Values" Method
The report method has node N report the value of its parameter, in protocol messages, to all the other nodes that need to hear it. IS-IS uses the report method. If the parameter is one that neighbors need to know, it is reported in a hello message (a message that does not get forwarded and is therefore seen only by the neighbors). If the parameter is one that all nodes (in an area) need to know, it is reported in an LSP. This method allows each node to have independent parameter settings and yet interoperate. For example, a node adjusts its Listen timer (when to declare a neighbor dead) for neighbor N based on N's reported Hello timer (how often N sends Hellos).
22.214.171.124 "Detect Misconfiguration" Method
Another method is the detect misconfiguration method, in which parameters are reported so that nodes can detect whether they are misconfigured. For example, the detect misconfiguration strategy makes sense when routers on a LAN might report to each other the (IP address, subnet mask) of the LAN.
An example in which the detect misconfiguration method is not the best choice is the OSPF protocol, which puts the Hello timer and other parameters into hello messages and has neighbors refuse to talk if the parameter settings aren't identical. This forces all nodes on a LAN to have the same Hello timer, but there might be legitimate reasons that the responsiveness versus overhead trade-off for one router might be different from that of another router so that neighbors might legitimately need different values for the Hello timer. Also, the OSPF method makes it difficult to change parameters in a running network because neighbors refuse to talk to each other while the network is being migrated from one value to another.
126.96.36.199 "Use My Parameters" Method
Another method is the use my parameters method. In one example (the bridge spanning tree algorithm), the Root bridge reports, in its spanning tree message, its values for parameters that should be used by all the bridges. In this way, bridges can be configured one by one, but a non-Root bridge simply stores its configured value in nonvolatile storage to be used if that bridge becomes Root. The values everyone uses for the parameters are the ones configured into the bridge that is currently acting as Root. This is a reasonable strategy provided that there is no reason to want nodes to be working with different parameter values.
Another example of this method is AppleTalk, in which the seed router informs the other routers of the proper LAN parameters, such as network number range. However, it is different from the bridge algorithm because if there is more than one seed router, they must be configured with the same parameter values.
A dangerous version of this method is one in which all nodes store the parameters when receiving a report. This might lead to problems because misconfiguring one node can cause all the other nodes to be permanently misconfigured. In contrast, consider the bridge algorithm. Although the Root bridge might get misconfigured with undesirable parameters, even if those parameters cause the network to be nonfunctional, simply disconnecting the Root bridge will cause some other bridge to take over and cause all bridges to use that bridge's parameter settings. Or simply reconfiguring the one Root bridge will clear the network.
The handshake protocol: This is bad enough when there are two people. Invariably I'm carrying things in both hands. Or I've just sneezed (see sneeze protocol). Or I've just come in from outside and my hand is embarrassingly cold (and I'm wishing the protocol were "stick your hand on the other person's neck protocol"). But when there's a group of n people all greeting each other, as with the wineglass protocol, it winds up as n2. And it takes complex choreography to avoid getting between a pair of handshakers.
To add to the confusion, when I go to a meeting such as an IETF meeting, most people look at least vaguely familiar. Some of them, apparently, I'm supposed to ignore. Others I'm supposed to acknowledge with a vague nod. To others I'm supposed to say "Hi." With others I'm supposed to shake their hands. Others I'm supposed to hug. Then there are some countries in which you're supposed to kiss even the most casual acquaintance on the cheek. And other countries in which you're supposed to kiss both cheeks. I've never figured out how people decide what protocol to use, and I'm always worried about offending people by guessing the wrong protocol.