A viral social media post claiming that Solana validators had fallen 84% reignited debate over the network’s decentralization over the weekend. The post framed SOL as a “centralized database,” turning a node-count dispute into a broader argument about the chain’s structure.
Solana co-founder Anatoly Yakovenko rejected that figure in a public response, saying validator participation was down by roughly 20% over the past 12 months, not 84%. He also said the discussion was blending two different categories: validators and full nodes. That distinction sits at the center of the argument.
Validator counts and full node counts are not the same
Yakovenko wrote that “Validators are not full nodes.” In his account, Solana runs around 5,000 full nodes, while Ethereum has about 8,300 full nodes. He added that Ethereum’s market capitalization is roughly four times that of Solana, suggesting that a raw comparison of node figures does not by itself prove network collapse or extreme centralization.
The issue matters because validator numbers are often used as a shorthand for decentralization. But validators responsible for consensus are different from full nodes handling chain verification and network data, and combining the two can distort how observers read the health of a blockchain.
Yakovenko ties the decline to the end of SFDP support
To explain the drop, Yakovenko pointed to the conclusion of the Solana Foundation Delegation Program (SFDP). The one-year bootstrap program covered voting costs for smaller validators, helping them stay active during an earlier stage of network growth. Once that support ended, some participants left, reducing validator participation.
That explanation did not end the criticism. Even after the 84% figure was challenged, skeptics kept pressing Solana on decentralization, and comparisons to a centralized database continued to circulate across social platforms.
Operating costs remain a central point of tension
Part of the debate has focused on how expensive it is to run Solana infrastructure. One social media user claimed that operating costs reached $20 million per validator, though that figure could not be independently verified. Industry reporting and comments from node operators describe a high-cost setup, but not a single confirmed number.
According to those reports, hardware expenses can start in the hundreds of dollars and rise into the thousands. Voting costs can push annual expenses into the tens of thousands of dollars. Some Solana validators reportedly stake millions of dollars in tokens and spend hundreds of thousands of dollars on operations. Several startups are also building tools aimed at letting users verify the network on consumer-grade hardware and home internet connections, but those products are still in alpha testing.
The latest dispute over Solana validator numbers has moved beyond one disputed percentage. It has brought old questions back into focus: what should count as decentralization, how much subsidy shaped participation, and how costly it remains to run serious network infrastructure.