Session 02
Infrastructure

PLATFORM
ARCHITECTURES.

Comparative analysis of consensus mechanisms, transaction models, and performance trade-offs across major Layer-1 protocols.

Core Objective

Analyze why blockchains perform differently. We will dissect the architectural decisions behind Ethereum, Solana, and Algorand, focusing on the "Blockchain Trilemma."

01

Layer-1 Protocols

Consensus
Proof of Stake (PoS)
Validators stake 32 ETH. Prioritizes decentralization.
Execution
EVM (Single Thread)
Sequential transaction processing.
Scaling
Rollup Centric
Offloads computation to Layer 2s.
Finality
~15 Minutes
Probabilistic finality until epochs finalize.
02

State Models

Account Model

Like a bank account. Global state maps addresses to balances. Used by Ethereum & Solana.

  • Stateful (easy for DApps)
  • Requires Nonce for ordering
  • Harder to parallelize

UTXO Model

Like cash. Unspent Transaction Outputs are consumed to create new ones. Used by Bitcoin & Cardano.

  • Stateless verification
  • Privacy benefits
  • Native parallel processing
03

Performance Benchmarks

Live Metrics Simulation
SYSTEM ACTIVE
Throughput Comparison (TPS)
Throughput (TPS)
ETH15-30
ALGO10k
SOL65k+
Time to Finality
ETH13 min
ALGO3.3 sec
SOL0.4 sec
Avg Cost ($)
ETH$2.50
ALGO$0.001
SOL$0.00025
04

The Trilemma Trade-off

PlatformOptimizationCompromise
EthereumSecurity & DecentralizationLow Throughput / High Cost
SolanaSpeed & ScalabilityHardware Centralization Risk
AlgorandEfficiency & FinalityPermissioned Relay Nodes

System Check

Architecture dictates performance limits

Account Model vs UTXO impacts developer experience

No "perfect" chain; only optimal trade-offs