You write to Cloud Bigtable with good row keys and want to monitor when to increase cluster size. Which two actions indicate scaling is needed? (Choose two.)

Sustained increases in write latency indicate that the current cluster size cannot handle the incoming write load efficiently, necessitating an increase in nodes to distribute the workload. Similarly, Bigtable clusters have a maximum storage capacity per node. When storage utilization approaches 70% or more, it signals that the cluster is nearing its storage limit and requires additional nodes to accommodate future data growth and maintain performance. Key Visualizer's Read and Write pressure indices are useful for identifying hot spots or uneven data distribution, not directly for determining when to scale the entire cluster based on a fixed threshold like 100. While read latency is important, a general threshold of 100ms isn't a universal indicator for scaling; it depends on application-specific SLAs.