Elrond — rebranded to MultiversX in November 2022 — is a sharded smart contract blockchain described in a 2019 whitepaper by Beniamin Mincu, Lucian Mincu, Lucian Todea, and the Elrond team. The whitepaper introduces Adaptive State Sharding: a sharding technique that simultaneously shards the network layer, the transaction processing layer, and the state storage layer, enabling linear throughput scaling as shards are added.

The consensus mechanism is Secure Proof of Stake (SPoS): validators are selected from each shard’s validator pool using a randomness source, then use BLS (Boneh-Lynn-Shacham) multisignature aggregation to sign blocks with a single compact signature — achieving fast consensus with minimal communication overhead.

> Whitepaper (PDF): elrond.com/files/Elrond_WP_v1.1.pdf

---

Publication and Context

Elrond was founded in 2017 and conducted an IEO (Initial Exchange Offering) on Binance Launchpad in July 2019, raising $3.25M. Mainnet launched June 2020.

By 2019, sharding was widely discussed as Ethereum’s long-term scaling path (Ethereum 2.0 sharding proposal). However, most proposals sharded only transaction processing — state and network connectivity remained unified. Elrond’s whitepaper argues that full state sharding (where each shard maintains only its portion of the global state) is necessary for true scalability.

The project’s real-world throughput claim: ~15,000 TPS on testnet with 3 shards; theoretically scaling to millions of TPS with more shards.

Rebranding: Elrond rebranded to MultiversX in November 2022, with the native token renamed from eGLD to EGLD (same acronym; primarily a brand change).

---

Adaptive State Sharding

Elrond’s sharding is described as “adaptive” because the number of shards adjusts dynamically based on network load:

Three Layers of Sharding

Layer	What is sharded	Purpose
Network sharding	Which validators communicate peer-to-peer	Reduce network bandwidth per validator
Transaction sharding	Which shard processes a transaction	Parallel transaction execution
State sharding	Which shard stores account state	Reduce per-node memory footprint

Bitcoin and Ethereum have none of these. Ethereum 2.0’s original sharding planned transaction+state sharding (later replaced by rollup-centric roadmap).

Meta Chain

Cross-Shard Transactions

1. Shard 1 deducts Alice’s balance and creates a cross-shard transaction receipt
2. The receipt is forwarded to Shard 2 in the next block cycle
3. Shard 2 credits Bob’s balance

Cross-shard transactions take 2 block cycles (~12 seconds with 6-second blocks). Optimizations reduce this in practice.

---

Secure Proof of Stake (SPoS)

SPoS addresses the security-vs-performance tradeoff in PoS consensus:

Validator Selection

• Selection is weighted by stake but uses VRF-like unpredictability to prevent validators from knowing they will be selected until the last moment (preventing targeted attacks)
• Validators are reassigned to different shards periodically to prevent collusion

BLS Multisignature

• Each validator signs independently; signatures are aggregated into a single BLS signature
• The aggregated signature verifies the approval of all signers simultaneously with a single pairing operation
• This is ~100× more compact than n individual ECDSA signatures

Block validity: ≥ 2/3 + 1 of the shard’s selected validators must sign.

---

Virtual Machines: WASM and EVM

MultiversX VM (WASM-based):

• Smart contracts compile to WebAssembly and run in a lightweight WASM VM
• Primary language: Rust (using the MultiversX Rust framework); C also supported
• WASM contracts are more efficient than EVM bytecode for CPU-bound computation

EVM Support (Ethereum integration):

• MultiversX added EVM compatibility to allow Solidity contracts to deploy without modification
• This dual-VM approach targets both WASM-native developers and existing Ethereum developers

---

Key Technical Properties

Property	Value
Consensus	SPoS (BLS multisig)
Sharding	Adaptive state sharding
Shard count	3 shards + Meta chain (mainnet)
Block time	~6 seconds
Finality	~12 seconds (2 blocks for intra-shard)
Smart contract VM	WASM + EVM
Primary language	Rust (WASM) / Solidity (EVM)
Token	EGLD

---

Reality Check

MultiversX delivered a working sharded blockchain on mainnet with real throughput numbers. BLS multisig and adaptive sharding are genuine technical implementations.

Challenges:

• Limited DeFi ecosystem: Despite technical sophistication, MultiversX’s DeFi ecosystem (xExchange as the primary DEX) remained significantly smaller than Ethereum, Solana, or BNB Chain.
• Cross-shard complexity deters developers: Building applications that span shards requires handling the 2-block cross-shard delay explicitly — many developers prefer simpler non-sharded environments.
• Developer activity: EGLD holder count and smart contract deployment rates suggest a smaller-than-anticipated developer ecosystem.
• Rebrand confusion: The Elrond → MultiversX rebrand (late 2022) coincided with the FTX collapse, reducing visibility during a poor market environment.

---

Legacy

MultiversX is one of few blockchains to implement full adaptive state sharding in production. Its BLS multisignature consensus is technically rigorous and its sharding approach influenced later academic discussions of modular blockchain design. The Sovereign Chain framework (2023) extends MultiversX to allow appchain deployment.

---

Related Terms

• Sharding
• Proof of Stake
• BLS Signatures
• WASM

---

Research

• Elrond Team. (2019). Elrond: A Highly Scalable Public Blockchain via Adaptive State Sharding and Secure Proof of Stake. elrond.com.

— Primary whitepaper; describes Adaptive State Sharding across three layers, SPoS with BLS, and cross-shard transaction flow.

• Boneh, D., Lynn, B., & Shacham, H. (2001). Short Signatures from the Weil Pairing. ASIACRYPT 2001.

— Original BLS signature paper; MultiversX’s block signing uses BLS aggregation directly derived from this work.

• Zamani, M., Movahedi, M., & Raykova, M. (2018). RapidChain: Scaling Blockchain via Full Sharding. ACM CCS 2018.

— Academic sharding protocol; relevant benchmark for comparing MultiversX’s full state sharding approach with other academic proposals.