Share
Solana’s infrastructure team Anza, published its 2026 technical blueprint on January 15, 2026, committing to three parallel protocol transformations that redefine consensus, transaction ordering, and throughput constraints. The roadmap positions Alpenglow consensus for Q3 mainnet deployment, ships Multiple Concurrent Proposers (MCPs) with in-protocol ordering enforcement, and removes bandwidth-latency bottlenecks through aggressive hardware and software optimization.
Anza's 2026 roadmap is here.
From Alpenglow to MCP, read the full blog from our CEO @bw_solana https://t.co/5urelPokWj
— Anza (@anza_xyz) January 15, 2026
New CEO Brennan Watt, who took the helm at Anza in January 2026, described the year as intensifying the focus on performance and network reliability while prioritizing improvements to the developer and user experience. The plan builds on 2025’s zero-downtime record and sub-400ms slot times, advancing from operational reliability into architectural redesign.
Validators approved Alpenglow with 98.27% support in September 2025, clearing the path for Solana’s largest consensus overhaul. The upgrade replaces Proof-of-History and TowerBFT with Votor and Rotor, targeting median finality of 150 milliseconds against the current 12.8-second window. Anza’s Q3 2026 timeline for mainnet deployment hinges on completing stress tests for network-fault scenarios and hardening validator reward mechanisms.
1/ Introducing the largest Solana Protocol change ever: Alpenglow, Solana's new consensus protocol conceived by the Anza Research team. Say goodbye to Tower BFT and Proof of History. Say hello to Votor & Rotor 🧵👇 pic.twitter.com/KPNQxQ1jBg
— Anza (@anza_xyz) May 19, 2025
The system moves voting off-chain, with validators submitting compact BLS-aggregated proofs after processing blocks. Anza noted that this approach under Alpenglow means that Solana can efficiently match traditional web infrastructure performance levels, and potentially unlock applications that demand real-time performance with cryptographic certainty.
Solana is getting a major upgrade and its called Alpenglow.
Why this upgrade and why does it matter?
– Solana is well known for its high TPS, but not many people know that a SOL transaction takes around 12.8 seconds to be considered fully permanent and irreversible.With… pic.twitter.com/ZKeAFRrut5
— Launchpad_Daddy (@Launchpad_Daddy) January 20, 2026
MCP introduces protocol-enforced transaction ordering at the replay stage, breaking leader monopoly on inclusion decisions. The mechanism ensures censorship resistance through multiple proposers while providing deterministic ordering that applications can leverage. Watt explained the 2026 version focuses on batch-level ordering, marking a fundamental change in how Solana and the broader crypto ecosystem handle market structure.
FULL TALK: @MaxResnick1 shows off the "final final final final final v2 final this time for real" design for MCP, the endgame for Solana block production 👀 pic.twitter.com/3dSx4GkrLW
— Solana (@solana) December 12, 2025
Anza’s bandwidth-latency framework attacks four independent bottlenecks. XDP shred transmission becomes the default path, bypassing kernel networking to dramatically increase Turbine capacity. Block limits expand to 100 million Compute Units via SIMD-0286, up from 60 million. Direct memory mapping reduces copy costs in the Solana Virtual Machine (SVM), while slot times drop below 400ms.
The biggest Solana news today wasn’t Grayscale or Western Union
It was @heliuslabs making 493 billion transactions searchable by account at Web2 speed
IBRL, from writes to reads 🏎️ https://t.co/iJcFcIJja6
— Solana (@solana) October 29, 2025
Transaction sending limits rise in Agave to match modern hardware, eliminating artificial constraints except during global congestion when stake-weighted throttling applies. The changes eliminate distinct and multiple constraints across networking, execution, memory management, and timing systems.
Scheduler bindings create separation between transaction scheduling logic and packing mechanics, establishing a controlled experimentation environment. Watt noted this reduces risks emanating from complex component dependencies that could cause validators to reject entire blocks before MCP’s protocol-level ordering arrives.
do you even IBRL bro? pic.twitter.com/AKI0JYxYWG
— Jito (@jito_sol) January 12, 2026
Developer-focused modifications include rent reduction, lowering long-term storage costs, larger transaction sizes through SIMD-0296, eliminating multi-step execution patterns, and p-ATA optimization, cutting costs for Associated Token Account interactions. Block Revenue Distribution via SIMD-0123 increases transparency in fee accounting and enables automatic reward redirection to infrastructure partners.
The simultaneous deployment of Alpenglow, MCP, and IBRL components positions 2026 as a structural transformation year for Solana’s scaling trajectory.
Share
