ERPC Upgrades Solana Shredstream Shared Endpoints. Co-located with Data Sources in Key Cities for Theoretically Fastest Performance
ERPC Upgrades Solana Shredstream Shared Endpoints. Co-located with Data Sources in Key Cities for Theoretically Fastest Performance
ELSOUL LABO B.V. (Headquarters: Amsterdam, the Netherlands; CEO: Fumitake Kawasaki) and Validators DAO have significantly upgraded the shared endpoints for Solana Shredstream.
From our extensive node operations, it has become even clearer that high‑clock servers deliver consistently strong performance across workloads. All shared endpoints have been switched to the highest‑clock configurations across every region.
In addition, the shared endpoints in Frankfurt, Amsterdam, and New York are now placed on the same network as the Shredstream data sources.
We care about single‑digit milliseconds. We continuously refine the platform to deliver the best possible environment, and this upgrade is part of that ongoing work.
Thank you as always for your support.
What changes with this upgrade
This upgrade achieves lower latency and better prevents data loss. Shredstream upstream reception uses UDP, and the longer the delivery distance, the greater the packet loss. Loss is problematic on its own and also slows continuous processing.
We minimized the distance of this UDP segment by co‑locating the shared endpoints directly under the data sources on the same network, reducing the source of loss to the minimum. As a result, the stability of both data and processing speed is expected to improve.
For distribution, gRPC runs over TCP with retransmission control. Combined with the zero‑distance upstream, this delivers stable high performance even on shared endpoints.
Applicable regions and operational notes
The high‑clock upgrade for shared endpoints is complete in the Chicago region as well. In Frankfurt, Amsterdam, and New York, the shared endpoints are co‑located on the same network as the data sources. The effect is most visible in workloads where differences of just a few milliseconds change outcomes.
Place your resources on the same network
If you pair the shared endpoints with EPYC VPS instances that sit on the same network, your application avoids round trips over the slower public internet. This further reduces end‑to‑end latency. Simply placing your application in close proximity removes the network bottleneck.
Bare metal servers remain the fastest option. While the premium line is currently sold out, we continuously add inventory. If you are interested, contact us via the Validators DAO official Discord. You can also join the waitlist.
- Validators DAO official Discord: https://discord.gg/C7ZQSrCkYR
How to get the new endpoints
In the Direct Shreds channel on the Validators DAO official Discord, re‑register from Whitelist IP to obtain the new shared Shredstream endpoints. They will appear immediately in the dashboard and you can switch over right away. If you have any questions, please contact us on Discord. There are no changes to specifications, rate limits, authentication, or price.
Problems ERPC solves
We build dedicated Solana infrastructure and a zero‑distance network to solve the following:
- Frequent transaction failures and latency variability common in generic RPC environments
- Performance throttling by many infrastructure providers
- The outsized impact of network distance on communication quality
- The difficulty for smaller projects to access high‑quality infrastructure
These improvements are the result of customer feedback and our research and development based on that feedback. We will keep this cycle running and continue on‑site verification and iteration to deliver the best possible environment at any given time.
For inquiries, trials, and waitlist requests:
- ERPC Official Site: https://erpc.global/en
- Validators DAO official Discord: https://discord.gg/C7ZQSrCkYR
Thank you for your continued support.
News
