Wormhole switching

Wormhole flow control, also called Wormhole switching or Wormhole routing is a system of simple flow control in computer networking based on known fixed links. It is a subset of flow control methods called Flit-Buffer Flow Control. [] . "Routing" defines the route or path taken to reach the destination. The wormhole technique does not dictate the route to the destination but decides when the packet moves forward from a router. Cut-through switching commonly called "Virtual Cut-through," operates in a similar manner, the major difference being that cut-through flow control allocates buffers and channel bandwidth on a packet level, while wormhole flow control does this on the flit level. In most respects, wormhole is very similar to ATM or MPLS forwarding, with the exception that the "cell" does not have to be queued.

Large network packets are broken into small pieces called flits (flow control digits). The first flit, called the header flit holds information about this packet's route (namely the destination address) and sets up the routing behavior for all subsequent flits associated with the packet. The head flit is followed by zero or more body flits, containing the actual pay load of data. The final flit, called the tail flit, performs some book keeping to close the connection between the two nodes. One thing special about wormhole flow control is the implementation of virtual channels.

A virtual channel holds the state needed to coordinate the handling of the flits of a packet over a channel. At a minimum, this state identifies the output channel of the current node for the next hop of the route and the state of the virtual channel (idle, waiting for resources, or active). The virtual channel m ay also include pointers to the flits of the packet that are buffered on the current node and the number of flit buffers available on the next node. (Dally and Towles 2004, p.237)

The name "wormhole" plays on the way packets are sent over the links: the address is so short that it can be translated before the message itself arrives. This allows the router to quickly set up the routing of the actual message and then "bow out" of the rest of the conversation. Since a packet is transmitted flit by flit, it may occupy several flit buffers along its path, creating a worm-like image. This, however, can be confusing since cut-through routing does the same thing.

Example

A wormhole flow control transmission may work as follows. Each node contains a router that will determine which path the packet will take through the network and holds the virtual channel state information:

1. A packet, P_a at an upstream node, say N₁, attempts to allocate an input virtual channel on a downstream node, N₂, to reach its destination, N₃. The input VC at each side of each node (call them N,S,E,W) will hold flit buffers and, in this case, will specify whether this input virtual channel is waiting, idle, or active. It will also specify which output virtual channel we are attempting to acquire. An output VC will hold only information about which input virtual channel it is reserved by.
2. P_a's header flit arrives at N₂'s West input VC, which happens to be in the idle state, so assuming we can buffer two flits, P_a's header flit and first body flit are buffered.
3. P_a wants to use N₂'s East output VC to reach N₃ so it specifies that in the VC state, but this output VC is currently being used by some other packet, P_b coming from the North. P_a is now "blocked", so the the West input VC on N₂ will enter the "wait" state. Note that N₂'s East output VC will specify that it is reserved by the North input VC. N₁ cannot send any more flits to N₂ now because the flit buffer is full.
4. P_b finishes transmitting and N₂'s East output VC becomes available.
5. P_a can now transmit to N₃ so the West input VC enters the "active" state, and the East output VC specifies that it is reserved by W.
6. P_a continues this transmission process until it reaches its destination.

Note that when P_a was blocked by P_b, the upstream node could not transmit any more packets downstream. This may extend upstream all the way to the source node as flit buffers fill up due to the blocking. This is an example of "backpressure".

Advantages

1. Wormhole flow control makes more efficient use of buffers than cut-through. Where cut-through requires many packets worth of buffer space, the wormhole method needs very few flit buffers (comparatively).

2. An entire packet need not be buffered to move on to the next node, increasing throughput.

3. Bandwidth and Channel allocation are decoupled

Wormhole techniques are primarily used in multiprocessor systems, notably hypercubes. In a hypercube computer each CPU is attached to several neighbours in a fixed pattern, which reduces the number of hops from one CPU to another. Each CPU is given a number (typically only 8-bit to 16-bit), which is its network address, and packets to CPUs are sent with this number in the header. When the packet arrives at an intermediate router for forwarding, the router examines the header (very quickly), sets up a circuit to the next router, and then bows out of the conversation. This reduces latency (delay) noticeably compared to store-and-forward switching that waits for the whole packet before forwarding. More recently, wormhole flow control has found its way to applications in Network On Chip systems (NOCs), of which multi-core processors are one flavor. Here, many processor cores, or on a lower level, even functional units can be connected in a network on a single IC package. As wire delays and many other non-scalable constraints on linked processing elements become the dominating factor for design, engineers are looking to simply organized interconnection networks, in which flow control methods play an important role.

An extension of worm-hole flow control is Virtual-Channel flow control, where multiple virtual channels are provided for each input port.

See also

*IEEE 1355
*SpaceWire

References