The XRP Ledger has achieved a significant reduction in node memory usage, recording improvements of up to 23% since the deployment of xrpld version 3.3.0. The update, released in August 2026, included a series of amendments, targeted bug fixes, and overall optimizations.
Technical improvements and root causes
RippleX developer Valentin Balaschenko detailed the process behind these optimizations, explaining how memory consumption was successfully reduced. According to Balaschenko’s official blog post, investigations began in late summer 2025 when node operators reported a steady increase in memory usage on their validators. Notably, these memory spikes occurred even on identically configured nodes and various hardware setups.
Analysis revealed that the issue was not a classic memory leak. Instead, the system’s memory allocator retained memory that the application itself had already freed. This atypical behavior led to increased resource consumption that did not improve after cache sweeps, which ordinarily clear expired entries.
Operators observed that node memory usage did not decline as expected, even after regular maintenance. The fix, included with the 3.3.0 update, specifically addressed how the memory allocator handled freed resources, resulting in significant efficiency gains.
A separate memory regression had been detected and rectified during the same period. However, this fix did not affect overall memory usage, since the underlying allocator issue persisted both before and after the patch, driving the need for a more comprehensive solution.
Community feedback and measured impact
Hussein Zangana, who serves as the director of community at the XRP Ledger Foundation, highlighted the tangible benefits in a recent post on X. Zangana reported, “Up to 23% memory savings on XRP Ledger nodes since 3.3.0. Given how expensive RAM is at the moment, this improvement is especially useful. Measured on (non/dUNL) validators as well as stock nodes, savings range from 13-23% depending on history.”
These savings offer a practical advantage for network participants, potentially lowering operational costs and improving performance for a wide range of node types.
Mini dictionary: XRP Ledger Foundation — An independent non-profit that supports the development, adoption, and governance of the XRP Ledger community.
Recent amendments: fixCleanup and fixCleanup3_4_0
On November 4, 2026, the XRP Ledger further strengthened its protocol with the activation of the fixCleanup amendment. This upgrade aimed to harden the core infrastructure and reinforce overall network stability.
In addition to fixCleanup, the fixCleanup3_4_0 amendment addressed a variety of key features and functionalities across the ledger. The changes target components including Single Asset Vaults, the Lending Protocol, Automated Market Makers, the permissioned decentralized exchange (DEX), trust lines, NFTs (non-fungible tokens), Multi-Party Transactions (MPTs), Escrow, Sponsorship, transaction signing, and permission delegation.
The fixCleanup3_4_0 update specifically modified the VaultClawback feature to reject pseudo-account holders, minimizing risks of asset mismanagement. Amendments to VaultDeposit, VaultWithdraw, and VaultClawback have improved the accuracy of asset rounding, ensuring that vault transactions deliver or receive precise amounts. VaultWithdraw now allows zero-asset withdrawals from impaired vaults for clearer accounting, and IOU issuers are exempt from their own global freeze on NFT offers in their respective currencies.
| Amendment | Main Focus | Key Improvement |
|---|---|---|
| fixCleanup | Node infrastructure | Protocol hardening and greater stability |
| fixCleanup3_4_0 | Core protocol features | Vaults, Lending, AMMs, Permissioned DEX, NFTs, Escrow, trust lines |
Outlook for the XRP Ledger
With these developments, both infrastructure operators and users of the XRP Ledger can expect more stable resource consumption and a more robust protocol, reflecting ongoing efforts to optimize performance and security.
Memory optimization in XRP Ledger nodes is expected to have a lasting impact, making validator and stock node operations more cost-efficient in a period of high RAM prices.
