starknet_committer: parallel reads using unordered futures

[nimrod-starkware]

Note

Low Risk
The change is additive and currently unused, with minimal impact beyond adding a new dependency; risk is mainly around future adoption of the concurrent code path and its correctness/perf characteristics.

Overview
Adds futures as a new dependency for starknet_committer and wires it into trie traversal.

Introduces an (currently unused) create_storage_tries_concurrently implementation that builds per-contract storage trie skeletons in parallel via FuturesUnordered, collecting results as they complete; existing behavior continues to use the sequential path.

^{Written by Cursor Bugbot for commit 096619a. This will update automatically on new commits. Configure here.}

[reviewable-StarkWare]

This change is 

[github-actions]

Artifacts upload workflows:

• Started at 2026-02-01T08:43:47Z
• Started at 2026-02-01T08:49:37Z
• Started at 2026-02-01T09:24:22Z
• Started at 2026-02-01T09:29:56Z
• Started at 2026-02-01T11:19:20Z
• Started at 2026-02-01T11:19:20Z
• Started at 2026-02-02T08:36:47Z
• Started at 2026-02-03T13:08:13Z
• Started at 2026-02-03T13:08:13Z
• Started at 2026-02-03T14:59:56Z
• Started at 2026-02-03T14:59:56Z
• Started at 2026-02-03T15:30:39Z
• Started at 2026-02-04T11:51:34Z
• Started at 2026-02-05T08:27:18Z
• Started at 2026-02-05T08:27:19Z
• Started at 2026-02-09T14:27:52Z
• Started at 2026-02-09T15:13:32Z
• Started at 2026-02-10T08:45:11Z
• Started at 2026-02-10T09:44:31Z
• Started at 2026-02-10T12:34:17Z
• Started at 2026-02-17T07:13:11Z
• Started at 2026-02-19T09:21:29Z
• Started at 2026-02-19T09:21:30Z
• Started at 2026-02-19T10:30:55Z

[nimrod-starkware]

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more

• starknet_patricia_storage: spawn blocking tasks for rocksDB reads #12168 : 2 dependent PRs (#12169 , #12172 )
• starknet_committer: parallel reads using unordered futures #12167 👈 (View in Graphite)
• starknet_committer: create sotrage tries sequentially #12166
• starknet_patricia,starknet_committer: dont pass storage mut when not needed #12165
• starknet_patricia_storage: non mut get in storage trait #12123 : 1 other dependent PR (#12252 )
• starknet_patricia_storage: dont cache reads in cached storage #12251
• starknet_patricia_storage: atomic counters for read stats #12122
• main-v0.14.2

This stack of pull requests is managed by Graphite. Learn more about stacking.

[chatgpt-codex-connector]

Codex usage limits have been reached for code reviews. Please check with the admins of this repo to increase the limits by adding credits.
Credits must be used to enable repository wide code reviews.

[ArielElp]

@ArielElp partially reviewed 3 files and made 3 comments.
Reviewable status: 0 of 3 files reviewed, 3 unresolved discussions (waiting on @nimrod-starkware).

---

crates/starknet_committer/src/db/trie_traversal.rs line 493 at r1 (raw file):

        // Prepare data for the async operation.
        let address = *address;

remove this line and use storage_tries.insert(*address, original_skeleton); at the end of the function

---

crates/starknet_committer/src/db/trie_traversal.rs line 494 at r1 (raw file):

        // Prepare data for the async operation.
        let address = *address;
        let cloned_storage = storage.clone();

We no longer need to clone, you can delete this line and use storage as-is below

---

crates/starknet_committer/src/db/trie_traversal.rs line 511 at r1 (raw file):

                &leaf_modifications,
                None,
                &address,

address

[ArielElp]

@ArielElp partially reviewed 2 files, made 1 comment, and resolved 3 discussions.
Reviewable status: 2 of 3 files reviewed, all discussions resolved (waiting on @nimrod-starkware).

[dorimedini-starkware]

@dorimedini-starkware reviewed 3 files and all commit messages, and made 4 comments.
Reviewable status: all files reviewed, 4 unresolved discussions (waiting on nimrod-starkware).

---

crates/starknet_committer/src/db/trie_traversal.rs line 0 at r3 (raw file):
without usage context, this is hard to review - you are rewriting existing trie-reading logic? copied from where?

---

crates/starknet_committer/src/db/trie_traversal.rs line 466 at r3 (raw file):

// TODO(Nimrod): Remove the `allow(dead_code)` once we use this function.
#[allow(dead_code)]
async fn create_storage_tries_concurrently<'a, Layout: NodeLayoutFor<StarknetStorageValue>>(

right?

Suggestion:

async fn create_storage_trie_original_skeletons_concurrently<

---

crates/starknet_committer/src/db/trie_traversal.rs line 490 at r3 (raw file):

        let trie_config = OriginalSkeletonTrieConfig::new_for_classes_or_storage_trie(
            config.warn_on_trivial_modifications(),
        );

this can be created outside the loop and cloned when needed, no?

Code quote:

        let trie_config = OriginalSkeletonTrieConfig::new_for_classes_or_storage_trie(
            config.warn_on_trivial_modifications(),
        );

---

crates/starknet_committer/src/db/trie_traversal.rs line 510 at r3 (raw file):

                address,
            )
            .await?;

makes it easier to read

Suggestion:

        // Create the future - tokio will poll all futures concurrently.
        futures.push(async move {
            let previous_leaves = None;
            let original_skeleton = create_original_skeleton_tree::<Layout::DbLeaf, Layout>(
                storage,
                contract_state.storage_root_hash,
                sorted_leaf_indices,
                &trie_config,
                &leaf_modifications,
                previous_leaves,
                address,
            )
            .await?;

[dorimedini-starkware]

@dorimedini-starkware reviewed 2 files and all commit messages.
Reviewable status: all files reviewed, 4 unresolved discussions (waiting on nimrod-starkware).

[nimrod-starkware]

@nimrod-starkware made 4 comments.
Reviewable status: all files reviewed, 4 unresolved discussions (waiting on dorimedini-starkware).

---

crates/starknet_committer/src/db/trie_traversal.rs line 466 at r3 (raw file):

Previously, dorimedini-starkware wrote…

right?

Yes, this is correct, but in the context of this file - we only build the original skeletons. So i think renaming it would be confusing.

---

crates/starknet_committer/src/db/trie_traversal.rs line 490 at r3 (raw file):

Previously, dorimedini-starkware wrote…

this can be created outside the loop and cloned when needed, no?

Done.

---

crates/starknet_committer/src/db/trie_traversal.rs line 510 at r3 (raw file):

Previously, dorimedini-starkware wrote…

makes it easier to read

Done.

---

crates/starknet_committer/src/db/trie_traversal.rs line at r3 (raw file):

Previously, dorimedini-starkware wrote…

without usage context, this is hard to review - you are rewriting existing trie-reading logic? copied from where?

Sorry, I agree.
The logic is copied from the sequential version. (see prev PR).
Somewhere up in the stack I have a PR that share the logic between these two flavors. (here)

[dorimedini-starkware]

@dorimedini-starkware reviewed 1 file and all commit messages, made 1 comment, and resolved 3 discussions.
Reviewable status: all files reviewed, 2 unresolved discussions (waiting on nimrod-starkware).

---

crates/starknet_committer/src/db/trie_traversal.rs line 502 at r5 (raw file):

        // Create the future - tokio will poll all futures concurrently.
        futures.push(async move {
            let prevoius_leaves = None;

typo

Code quote:

let prevoius_leaves = None;