Making Solana fast: from 33% skip rates to 100M compute unit blocks
Alessandro Decina from Anza joins Pirates Parley to break down how the Agave validator client went from constant congestion and 33% skip rates to 100 million compute-unit blocks, 200-millisecond slots, and the Alpenglow consensus rewrite.
Alessandro spent 15 years in open-source multimedia and audio-video streaming, wrote one of the early XDP packet-processing implementations in C, and then added eBPF support to the Rust compiler.
That work pulled him into Solana, where he now leads performance at Anza on Agave.
They cover:
Skip rate falling from 33% to 0.5% by version 2.3, and outages becoming a non-event The TPU rewrite to QUIC, stake-weighted QoS and why early spam still got through Why one syscall per packet caps throughput, and how XDP writes straight to the network card The invalidator is an internal team that attacks the cluster to surface the next bottleneck Rewriting parts of the Turbine with XDP so that much larger blocks can propagate Block size moving from 60 million to 100 million compute units, with headroom for more Shorter 200 millisecond slots targeted for the 4.2 release around the end of July Shrinking the leader ingest window from 30 seconds to 2.5 seconds with help from RPC providers Why Alpenglow matters most for deleting years of fragile consensus code A candid take on whether running two clients really improves security The Firedancer rivalry and the offsite that made him commit to making Agave faster His five-year vision of open source, permissionless global finance
Triton One: triton.one/pricing Alessandro Decina: x.com/alessandrod Anza: anza.xyz
Making Solana fast: from 33% skip rates to 100M compute unit blocks
Bringing privacy to everything on Solana: Arcium
The web janitor of the Solana Foundation on agents, AEO and building on Solana
More from Pirates Parley
Velocity: Inside the Drift Rebuild, North Korea's Hack, and What "Good OPSEC" Actually Means
Noah Prince is back on Pirates Parley — this time as the protocol engineer rebuilding what was Drift into Velocity, after North Korea drained ~$290M through a socially-engineered admin multisig.We get into how the hack actually worked (durable nonces, compromised laptops, and a circuit-breaker bypass that shouldn't have existed), what real OPSEC looks like when Ledgers can't clear-sign, why "seven years to pay users back" is a failure state, and the new tech Velocity is shipping post-launch: prop AMMs on perps and dynamic on-chain routing.Also: Tether's rescue deal, how Cindy negotiated it, the DFX claim mechanism, why Noah stays bullish on Solana ("Bitcoin and Ethereum are so shitty"), and — for the traders — how to actually model your loss exposure before depositing anywhere.Velocity is in closed beta. Talk to Tracy (@TracyBBD) to get on the whitelist. They're also hiring smart-contract engineers.0:00 Welcome — Noah's back after the Helium turducken2:05 How Drift got hacked: North Korea, ~$290M, the multisig4:30 How Noah got the call from Jacob Creech6:05 Tether's rescue and the DFX claim mechanism11:15 Why relaunch — winning Solana's perp war15:40 OPSEC deep dive: durable nonces and the anatomy of the exploit19:35 Hot take: Ledgers are worthless doorstops22:25 The fixes: circuit breakers, tiered admins, time locks26:25 Auditing culture and why you diversify across protocols33:55 Noah's Solana origin story — $8 in Ethereum fees40:35 Seven years or bust: what "success" actually means48:25 Justin Sun, team security, and modeling your real DeFi risk57:10 Leverage, launch timeline, and what's coming (prop AMMs, routing)1:09:45 Beta access, hiring, and Drift's governance risks1:16:15 Brother Jordan's Imperial + what's next on the pod
Sunsetting the mempool: Lucas Bruder on Jito, BAM, and betting long-term on Solana
Lucas Bruder, co-founder and CEO of Jito Labs, joins Pirates Parley to break down why they shut off the public mempool, how BAM works, and what confidential block building on Solana looks like inside a secure enclave.Lucas got here via Tesla body controls on the Model X, an Xbox controller for surgical robots, a 50,000 pound bulldozer at Built Robotics, and lidar firmware at Ouster, before falling down the MEV rabbit hole and starting Jito in 2021.They cover:- Lucas's path from Tesla and Built Robotics to lidar and then Jito- how the Jito block engine works, and why the best analogy is Cloudflare- the peak day the Trump token launched, when Jito generated the network $20m and made up more than 50% of Solana REV- why they sunset the public mempool, and why Steve thinks it's the biggest pivot any blockchain company has made- BAM: transparent, verifiable, confidential block building inside a secure enclave- the maker priority plugin, and BAM running on just under 33% of stake- Jump running Bison as the number one prop amm on Solana- what has to happen for tokenised securities to work: Clarity, an SEC exemption, and users- why Jito Labs built JTX and how they think about execution- restaking, the honest reflection, and where Jito is focusing now- 100m CU blocks activated last week and the road to 200ms slots and Alpenglow- advice for new builders coming into Solana--------------------------------------------------------------00:00 - Welcome and cold open01:14 - Lucas's path: Tesla, hardware consultancy, cadavers05:13 - Built Robotics and the bulldozer06:46 - Ouster, MEV, and falling down the rabbit hole11:12 - What Jito does today: the market layer16:40 - The Jito block engine explained21:11 - The mempool: what it was, where it went wrong24:49 - Sunsetting the mempool: the biggest pivot in blockchain28:05 - BAM: block assembly marketplace31:10 - Maker priority, 33% stake, and consistent execution36:41 - Tokenised securities, Clarity, and what has to happen46:34 - Why Jito built JTX48:52 - Team size, focus, and letting restaking run54:11 - 100m CU blocks, Alpenglow, and shipping at Anza speed1:04:47 - Advice for new builders
Streamflow: solution first, problem later, and the alpha on equity-backed tokens
Malisha Stanojevic, founder and CEO of Streamflow, joins Pirates Parley to walk through five years of building the token distribution platform on Solana, and to drop the news on what they are turning it into next.Malisha joined Solana in May 2021 via a hackathon, built a Sablier-inspired token streaming protocol before he knew what problem it solved, and raised his first check at the Breakpoint 2021 closing party by typing a note on his phone because it was too loud to talk. Streamflow has since had over 40,000 tokens go through its programs and peaked at two billion dollars worth of tokens in escrow.They cover:Building the solution before finding the problem, and why he does not recommend itThe first check raised on a phone screen at Breakpoint 2021A 15-person VC call that turned into an interrogationSolana Ventures backing Streamflow and two of its competitors at the same timeWhy ETH, Aptos and Sui conferences felt like a different planetThe $1-a-day customer stream that got topped up with $15,000What it feels like to sit on two billion in customer escrowStride, Asymmetric Research, and how AI is now inside their CI/CD and code reviewWhy mental health is a founder virtue, not a weaknessThe alpha: equity-backed tokens, a Cayman Islands legal framework, and Streamflow becoming an onchain capital formation platformThe STREAM token transition and the Solana Summit Belgrade rollout00:00 - welcome and intro00:30 - how Malisha found Solana via a hackathon in 202103:40 - solution first, problem later: the Streamflow origin story08:50 - raising the first check on a phone screen at Breakpoint 202113:55 - the 15-person VC call about stolen open-source code18:20 - bias for action and why perfect is not that important21:20 - ETH vs Solana culture, and Solana Ventures backing three competitors26:20 - what Streamflow actually does: vesting, locks, airdrops, staking29:20 - the $1-a-day topup that extended a stream by 11 years33:20 - two billion in escrow and the top two fears in life37:40 - Stride, Asymmetric Research, and AI in the security stack42:40 - mental health as a founder virtue48:00 - the numbers: 20 people, $2.5M in 2025, still profitable50:20 - the alpha: equity-backed tokens and onchain capital formation56:00 - Cayman Islands framework and the STREAM transition58:00 - Solana Summit Belgrade, and next week on Pirates Parley