Walrus Latency. Figure 8 illustrated the end-to-end latency experienced by the client. We start measuring before the client encodesthe blob and finish when it observes a proof-of-availability confirmation on the blockchain. Each point represents the p50 over 5

minutes of runs; error bars indicate p90.
The graph shows that read latency remains low, even for large
blobs. For small blobs (less than 20 MB), the latency stays below 15
seconds. For large blobs (130 MB), the latency increases to around
30 seconds

{future}(WALUSDT)