#Walrus
@Walrus 🦭/acc
Triggering the Change process in the Walrus Protocol is a critical governance and operational event that facilitates the transition between "epochs." This process ensures the network remains dynamic, secure, and capable of adapting to changes in node participation and stake distribution. Because Walrus relies on the Sui blockchain for its control plane, this transition is a highly coordinated smart contract execution.
1. The Epoch Transition
The lifecycle of the Walrus network is divided into fixed periods called epochs. At the end of each epoch, the protocol must "trigger a change" to refresh the Active Storage Committee. This is necessary to account for new WAL tokens being staked, delegators moving their support between nodes, or underperforming nodes being removed. The change is officially initiated through a system-level call on the Sui blockchain, often referred to as initiate_epoch_change.
2. Stake Re-Weighting and Committee Selection
When the change is triggered, the protocol takes a "snapshot" of all WAL token delegations.
Ranking: All candidate nodes are ranked based on their total stake. Only a top-tier group of nodes (the "Active Set") is selected to handle the data plane for the next epoch.
Probability Weighting: The process uses these stake weights to determine how many "shards" or "slivers" of data each node is responsible for. This ensures that nodes with more "skin in the game" carry a proportional share of the storage load.
3. Price Discovery and Commitment
A unique aspect of triggering the change in 2026 is the Stake-Weighted Price Discovery. During the transition, selected nodes must submit their storage price bids for the upcoming epoch. The protocol aggregates these bids and selects a uniform market price (typically the 66.67th percentile of the total stake). This ensures that as the epoch changes, the price for storing new "blobs" remains competitive and reflective of the current supply and demand for decentralized storage.
4. Handling Node Churn and Data Continuity
The most complex part of the change process is ensuring Data Availability remains uninterrupted. As the committee changes, some nodes may leave the active set while others join. The protocol manages the secure "handover" of metadata and ensures that the Red Stuff erasure coding continues to function correctly. This seamless transition prevents any "dark periods" where data might become unreachable during the reconfiguration, maintaining the 2026 standard of 99.99% availability for enterprise-grade AI and multimedia datasets.