Over the past weeks, we have been actively optimizing LocalShred™ across our Amsterdam network. The goal was simple: reduce the time between receiving Solana shreds and delivering them directly to client servers as much as possible.
Last month, Jito ShredStream improved its Amsterdam performance by roughly 4ms, which was a strong improvement for the ecosystem. At that point, LocalShred™ was still around 1ms faster, but we did not feel that was enough.
After further work on routing, local delivery paths, packet handling, and internal distribution, the first new results are now showing LocalShred™ averaging around 4 to 5ms faster than Jito ShredStream from our Amsterdam network.
For Solana infrastructure, this matters. Solana shreds are one of the earliest signals your server can receive from the network. When your RPC node, validator, or trading system receives shreds earlier, it can start processing earlier. In latency sensitive environments, even a few milliseconds can make a meaningful difference.
What LocalShred™ brings to DedicatedNodes clients
This is not a one time improvement. We are continuing to tune the system, improve routing, and measure real world performance so our clients receive the best possible shred delivery from our network.
You can read more about LocalShred™ here:
https://www.dedicatednodes.io/low-latency-solana-shreds/
We are excited about the progress so far, and even more excited about what comes next.
