Why
In 2026 a routing error on Solana took roughly 29% of stake offline at once and the network came within a few percentage points of losing finality. Nothing was hacked and no key was stolen — the consensus math worked exactly as designed. The failure was that a single infrastructure fault could reach a third of the validator set at the same time, which is a question about network topology, not cryptography. The interesting output is a number nobody publishes: for a given chain, what is the largest slice of stake that shares one correlated point of failure, and how does that compare to the fraction that halts finality (33% for both Solana's and Ethereum's BFT thresholds)?
How it works
Planned as a measurement, not an essay: pull the active validator set and its stake weights from public RPC, resolve each advertised gossip/TPU endpoint to an IP, map those to ASN and hosting provider with a public IP-intelligence dataset, then aggregate stake by ASN, by provider, and by geographic region. The headline figure is the stake-weighted Herfindahl index plus the single largest correlated bucket, both plotted against the 33% halt line. The same pipeline runs unchanged against Ethereum's beacon chain for comparison, where the concentration hides one layer down — in staking pools and the handful of clouds their operators rent from rather than in the validator count itself.
Where it lands in Jayverse
- Devnet: measure correlated infrastructure, not just node count, when Devnet moves to an OP-Stack L2. For any future sequencer or validator setup, aggregate stake by ASN and hosting provider against the halt threshold, the same pipeline this card proposes for Solana and Ethereum.
- Auditor: add a standing check for the largest correlated-failure bucket on any chain Jayverse depends on. Track what fraction of stake or sequencing power sits behind one ASN or provider on Sepolia today and any future L2, and how close that sits to the finality-halt line.
Key expressions
| Expression | 뜻 · 쓰이는 자리 |
|---|---|
| come within (a few points) of | ~포인트 차이로 (거의) ~할 뻔하다 · 위기를 아슬아슬하게 피했을 때. "came within a few percentage points of losing finality" |
| correlated point of failure | 상관된 단일 장애 지점 · 여러 참여자가 동시에 함께 무너지는 지점. "one correlated point of failure could reach a third" |
| headline figure | 대표 수치, 핵심 지표 · 보고서에서 가장 먼저 눈에 띄는 숫자. "The headline figure is the stake-weighted Herfindahl index" |
| hide one layer down | 한 단계 아래 숨어 있다 · 겉으로 드러나지 않고 더 깊은 구조에 위험이 있을 때. "the concentration hides one layer down" |
| aggregate by | ~기준으로 (데이터를) 집계하다 · 여러 항목을 특정 범주로 묶어 합산할 때. "aggregate stake by ASN, by provider" |
| unit of independence | 독립성의 단위(비유) · 무엇을 기준으로 '진짜 분산되어 있는지' 판단해야 하는지. "the real unit of independence" |
| routing error | 라우팅 오류, 네트워크 경로 설정 실수 · 해킹 없이도 대규모 장애를 일으킬 수 있는 원인. "a routing error on Solana took roughly 29% of stake offline" |
| ASN | 자율시스템 번호(Autonomous System Number) · 인터넷 라우팅상 독립 네트워크를 식별하는 번호, 상관 장애의 단위로 쓰임. "map those to ASN and hosting provider" |
| BFT | 비잔틴 장애 허용(Byzantine Fault Tolerance) · 합의가 깨지는 정족수 기준(33%)을 가리킬 때 쓰는 용어. "Solana's and Ethereum's BFT thresholds" |
| TPU | 트랜잭션 처리 유닛(Transaction Processing Unit) · 솔라나 검증자가 트랜잭션을 받는 네트워크 엔드포인트. "resolve each advertised gossip/TPU endpoint to an IP" |