DLTs require a mechanism to control the rate of transactions entering the network, as the limited resources of nodes (e.g., bandwidth, CPU) may become overwhelmed, leading to low throughput. This is similar to a highway, where a large number of vehicles entering the flow of traffic can lead to congestion.
Currently, the Tangle implements a simple spam protection mechanism based on PoW. However, controlling the rate of transaction issuance by PoW alone is impractical, as it introduces mining races. We propose an adaptive rate control mechanism which intelligently varies the difficulty of the PoW per node based on different factors, such as number of recently-issued transactions and mana.
It is important to note that the use of PoW as a rate control mechanism is unrelated to PoW-based consensus, which our Coordicide approach explicitly eliminates. Our rate control mechanism is designed in such a way that it does not lead to mining races, nor does it require a large amount of energy.
In our model, nodes with higher amounts of mana have the ability to issue more transactions, without the same PoW requirements as low-reputation nodes. Regardless of a node’s mana, PoW difficulty increases with transaction rate. i.e. in order to issue more transactions in a short time interval, a node must increase the difficulty of the cryptographic puzzle, while for lower transaction rates a much lower difficulty would be sufficient.
We enforce a maximum transaction rate per node to further prevent spam. This system brings a double benefit:
- It protects the network against malicious fast actors, e.g., nodes using ASICs to spam the network and/or influence the consensus.
- IOTA’s main use case is IoT, which consists of nodes with limited resources that are likely to issue transactions at a slow rate. This mechanism allows slower nodes to successfully propagate their transactions through the network.