chore(ci): Run benchmarks on Earthly

[spalladino]

Migrates benchmark jobs from CCI to Earthly+GA.

Introduces a new yarn-project/scripts/Earthfile with jobs for downloading logs, aggregating them, and commenting, which are called from a new bench-summary job in GA's CI workflow that runs after e2e. Adds a new UPLOAD_LOGS function in the yarn-projects/end-to-end/Earthfile to upload logs to S3, gets called from all bench e2e jobs.

Perhaps we can ditch S3 for storing logs (not historic benches, since those are needed for comparisons) and use Earthly artifacts for that, but we can do that later down the road.

[AztecBot]

Benchmark results

Metrics with a significant change:

• circuit_simulation_time_in_ms (merge-rollup): 5,388 (-16%)
• tx_pxe_processing_time_ms (0): 1,733 (+25%)
• batch_insert_into_indexed_tree_20_depth_ms (512): 462 (+23%)
• tx_sequencer_processing_time_ms (2): 1,009 (+28%)
• tx_sequencer_processing_time_ms (4): 1,180 (+27%)
• tx_sequencer_processing_time_ms (8): 1,410 (+26%)
• l2_block_public_tx_process_time_in_ms (8): 10,546 (+19%)

Detailed results

All benchmarks are run on txs on the Benchmarking contract on the repository. Each tx consists of a batch call to create_note and increment_balance, which guarantees that each tx has a private call, a nested private call, a public call, and a nested public call, as well as an emitted private note, an unencrypted log, and public storage read and write.

This benchmark source data is available in JSON format on S3 here.

L2 block published to L1

Each column represents the number of txs on an L2 block published to L1.

Metric	8 txs	32 txs	64 txs
l1_rollup_calldata_size_in_bytes	740	740	740
l1_rollup_calldata_gas	6,668	6,692	6,656 (-1%)
l1_rollup_execution_gas	586,800	586,824	586,788
l2_block_processing_time_in_ms	1,619 (+8%)	5,170 (-5%)	9,767 (-5%)
l2_block_building_time_in_ms	25,779 (+15%)	90,753 (+2%)	168,169 (-5%)
l2_block_rollup_simulation_time_in_ms	25,496 (+15%)	89,851 (+2%)	166,706 (-5%)
l2_block_public_tx_process_time_in_ms	⚠️ 10,546 (+19%)	37,665 (+9%)	69,594 (-1%)

L2 chain processing

Each column represents the number of blocks on the L2 chain where each block has 16 txs.

Metric	5 blocks	10 blocks
node_history_sync_time_in_ms	14,788 (-8%)	27,045 (-9%)
node_database_size_in_bytes	20,766,800	37,396,560
pxe_database_size_in_bytes	29,868	59,425

Circuits stats

Stats on running time and I/O sizes collected for every circuit run across all benchmarks.

Circuit	circuit_simulation_time_in_ms	circuit_input_size_in_bytes	circuit_output_size_in_bytes
private-kernel-init	192 (+10%)	46,819	28,822
private-kernel-inner	225 (+11%)	76,734	28,822
private-kernel-ordering	251 (+11%)	53,040	42,751
base-parity	77.6 (+16%)	128	64.0
base-rollup	14,780 (-10%)	111,062	925
root-parity	1,658 (+8%)	27,064	64.0
root-rollup	75.3 (+17%)	11,454	789
public-kernel-app-logic	2,027 (+3%)	54,321	22,677
public-kernel-tail	4,951 (+2%)	159,770	3,898
merge-rollup	⚠️ 5,388 (-16%)	2,696	925
public-kernel-teardown	169 (+5%)	54,321	30,042
public-kernel-setup	223 (+1%)	54,321	30,042

Tree insertion stats

The duration to insert a fixed batch of leaves into each tree type.

Metric	1 leaves	16 leaves	64 leaves	128 leaves	512 leaves	1024 leaves	2048 leaves	4096 leaves	32 leaves
batch_insert_into_append_only_tree_16_depth_ms	11.7 (+8%)	20.1 (+13%)	N/A	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_count	16.7	31.8	N/A	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_16_depth_hash_ms	0.681 (+7%)	0.614 (+12%)	N/A	N/A	N/A	N/A	N/A	N/A	N/A
batch_insert_into_append_only_tree_32_depth_ms	N/A	N/A	54.2 (+5%)	83.2 (+4%)	300 (+15%)	486 (-2%)	1,048 (+7%)	1,900 (-2%)	N/A
batch_insert_into_append_only_tree_32_depth_hash_count	N/A	N/A	95.9	159	543	1,055	2,079	4,127	N/A
batch_insert_into_append_only_tree_32_depth_hash_ms	N/A	N/A	0.553 (+5%)	0.513 (+4%)	0.544 (+14%)	0.454 (-3%)	0.495 (+6%)	0.455 (-2%)	N/A
batch_insert_into_indexed_tree_20_depth_ms	N/A	N/A	63.8 (+6%)	121 (+2%)	⚠️ 462 (+23%)	716 (-2%)	1,529 (+4%)	2,865 (-2%)	N/A
batch_insert_into_indexed_tree_20_depth_hash_count	N/A	N/A	105	208	692	1,363	2,707	5,395	N/A
batch_insert_into_indexed_tree_20_depth_hash_ms	N/A	N/A	0.557 (+5%)	0.543 (+1%)	0.624 (+22%)	0.493 (-2%)	0.527 (+4%)	0.499 (-2%)	N/A
batch_insert_into_indexed_tree_40_depth_ms	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A	67.3 (+1%)
batch_insert_into_indexed_tree_40_depth_hash_count	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A	108
batch_insert_into_indexed_tree_40_depth_hash_ms	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A	0.594 (+1%)

Miscellaneous

Transaction sizes based on how many contract classes are registered in the tx.

Metric	0 registered classes	1 registered classes
tx_size_in_bytes	44,613	535,433

Transaction size based on fee payment method

Metric	native fee payment method	fpc_public fee payment method	fpc_private fee payment method
tx_with_fee_size_in_bytes	897	1,145	1,345

Transaction processing duration by data writes.

Metric	0 new note hashes	1 new note hashes	2 new note hashes
tx_pxe_processing_time_ms	⚠️ 1,733 (+25%)	1,106 (+11%)	4,173 (+14%)

Metric	1 public data writes	2 public data writes	3 public data writes	4 public data writes	5 public data writes	8 public data writes
tx_sequencer_processing_time_ms	907 (+3%)	⚠️ 1,009 (+28%)	1,638 (+15%)	⚠️ 1,180 (+27%)	2,498 (+14%)	⚠️ 1,410 (+26%)