DataStore/orca-whirlpool-historical-data

Orca Whirlpools Historical Data

Decoded Solana mainnet instructions and events from Orca Whirlpools, Orca's concentrated-liquidity AMM: swaps, position open/close, liquidity changes, fee and reward collection, and the Token-2022 (_v2) instruction family.

54 tables, 37,021 rows, one row per decoded instruction or event.

This is a free sample from datastore.sh, which publishes the complete history as versioned Parquet.

Read this before you analyse it

The sample is capped per table, not per time window. 34 of the 54 tables stop at exactly 1,000 rows. Each table therefore covers however long it took to reach the cap, and those windows differ by orders of magnitude:

Table	Rows	Time span to fill
orca_whirlpool_swap	1,000	52 seconds
orca_whirlpool_swap_v2	1,000	46 seconds
orca_whirlpool_two_hop_swap	1,000	2 days 13 hours
orca_whirlpool_two_hop_swap_v2	1,000	6 days 3 hours

That spread is itself the finding — single-hop swaps arrive roughly 4,000× faster than two-hop ones. But it means row counts across tables are not comparable as activity measures, and the data won't support volume trends or backtests.

v1 and v2 are concurrent, not sequential. The _v2 instructions are the Token-2022 variants, live alongside v1 rather than replacing it. Both fill 1,000 rows in under a minute of the same blocks. Analyse them together or you see half the market.

Amounts are raw base units. Divide by 10^decimals of the mint. Nothing is decimal-adjusted.

a_to_b is the direction flag, and amount_specified_is_input tells you whether amount is the exact input or the exact output — read it before comparing amount across rows.

What the sample actually contains

Comparing the two swap families over near-identical blocks:

	swap (v1)	swap_v2 (Token-2022)
Rows	1,000	1,000
Distinct whirlpools	73	136
Columns	24	29
a→b / b→a	896 / 104	815 / 185
Exact-in swaps	97.6%	96.4%

v2 touches nearly twice as many distinct pools (136 vs 73) in the same span. The Token-2022 path is where the long tail of newer mints trades; v1 flow concentrates in a smaller set of established pools. Both are overwhelmingly exact-in (~97%), so exact-out routing is rare on Orca directly.

Columns shared by every table

Column	Meaning
ts	Block time, UTC
slot	Solana slot
tx_index	Position of the transaction in the block
ix_path	Instruction position; more than one index means an inner (CPI) instruction
sig	Transaction signature, base58 — the join key across tables
remaining	Remaining accounts, JSON, pubkeys as byte arrays (convert to base58 to match explorers)

Order events within a transaction by slot, tx_index, then ix_path.

Loading

from datasets import load_dataset

ds = load_dataset("DataStore/orca-whirlpool-historical-data", "orca_whirlpool_swap_v2")

Each table is a separate config named after its file. With pandas:

import pandas as pd

url = "https://huggingface.co/datasets/DataStore/orca-whirlpool-historical-data/resolve/main/orca_whirlpool_swap_v2.csv"
df = pd.read_csv(url)

Worth trying

Check whether the v1/v2 pool split is really about token standard by joining pool addresses across the two tables — how many whirlpools appear in both? Use open_position against close_position to estimate how long concentrated positions stay open. Compare collect_fees to collect_fees_v2 to see whether Token-2022 pools return more fees per position.

Tables

Table	Rows	Columns
orca_whirlpool_close_bundled_position	1,000	14
orca_whirlpool_close_position	1,000	14
orca_whirlpool_close_position_with_token_extensions	1,000	14
orca_whirlpool_collect_fees	1,000	17
orca_whirlpool_collect_fees_v2	1,000	22
orca_whirlpool_collect_protocol_fees_v2	1,000	21
orca_whirlpool_collect_reward	1,000	16
orca_whirlpool_collect_reward_v2	1,000	19
orca_whirlpool_decrease_liquidity	1,000	22
orca_whirlpool_decrease_liquidity_v2	1,000	27
orca_whirlpool_increase_liquidity	1,000	22
orca_whirlpool_increase_liquidity_by_token_amounts_v2	1,000	25
orca_whirlpool_increase_liquidity_v2	1,000	27
orca_whirlpool_initialize_dynamic_tick_array	1,000	14
orca_whirlpool_initialize_pool	1,000	23
orca_whirlpool_initialize_pool_v2	1,000	24
orca_whirlpool_initialize_pool_with_adaptive_fee	1,000	26
orca_whirlpool_liquidity_decreased_event	1,000	15
orca_whirlpool_liquidity_increased_event	1,000	15
orca_whirlpool_liquidity_repositioned_event	1,000	24
orca_whirlpool_open_bundled_position	1,000	19
orca_whirlpool_open_position	1,000	22
orca_whirlpool_open_position_with_metadata	1,000	26
orca_whirlpool_open_position_with_token_extensions	1,000	21
orca_whirlpool_pool_initialized_event	1,000	16
orca_whirlpool_position_opened_event	1,000	10
orca_whirlpool_reposition_liquidity_v2	1,000	31
orca_whirlpool_reset_position_range	1,000	16
orca_whirlpool_swap	1,000	24
orca_whirlpool_swap_v2	1,000	29
orca_whirlpool_traded_event	1,000	16
orca_whirlpool_two_hop_swap	1,000	35
orca_whirlpool_two_hop_swap_v2	1,000	40
orca_whirlpool_update_fees_and_rewards	1,000	12
orca_whirlpool_initialize_token_badge	669	15
orca_whirlpool_set_fee_rate	616	12
orca_whirlpool_initialize_position_bundle	478	17
orca_whirlpool_initialize_tick_array	445	13
orca_whirlpool_delete_position_bundle	281	14
orca_whirlpool_lock_position	199	18
orca_whirlpool_transfer_locked_position	70	16
orca_whirlpool_set_collect_protocol_fees_authority	54	11
orca_whirlpool_set_fee_authority	54	11
orca_whirlpool_set_reward_emissions_super_authority	54	11
orca_whirlpool_collect_protocol_fees	52	16
orca_whirlpool_set_protocol_fee_rate	17	12
orca_whirlpool_set_reward_emissions_v2	11	13
orca_whirlpool_initialize_fee_tier	7	15
orca_whirlpool_set_default_protocol_fee_rate	6	11
orca_whirlpool_initialize_reward_v2	4	18
orca_whirlpool_initialize_config_extension	1	13
orca_whirlpool_set_config_extension_authority	1	12
orca_whirlpool_set_token_badge_attribute	1	14
orca_whirlpool_set_token_badge_authority	1	12

License & citation

CC BY-SA 4.0 — share and adapt, including commercially, with attribution and alike licensing.

datastore.sh (2026). Orca Whirlpools Historical Data (Sample). https://datastore.sh/datasets/orca-whirlpool