eip-4844-sim.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "# EIP-4844 Fee Market Analysis: Simulation and Backtest"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Description"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "Proto-danksharding or [EIP-4844](https://eips.ethereum.org/EIPS/eip-4844) introduces a new transaction format for “blob-carrying transactions” which contain a large amount of data that cannot be accessed by EVM execution, but whose commitment can be accessed. The goal is to facilitate an ecosystem-wide move to rollups: Optimism and Arbitrum provide fees that are ~3-8x lower than the Ethereum base layer, and ZK rollups, which have better data compression and can avoid including signatures, have fees ~40-100x lower than the base layer.\n",
    "\n",
    "The long-term scaling solution is still data sharding: add ~16 MB per block of dedicated data space to the chain that rollups could use. This EIP is an intermediate step which implements (1) same transaction format as sharding and (2) creates the gas fee market for this type of transactions, but has the following differences:\n",
    "\n",
    "- No sharding of transactions: the data from blob transactions is fully downloaded by all consensus nodes (but can be deleted after only a relatively short delay)\n",
    "\n",
    "- Reduced cap on the number of these transactions that can be included: target of ~0.25 MB per block and limit of ~0.5 MB\n",
    "\n",
    "## Data Gas Fee Market\n",
    "\n",
    "Data gas is introduced as a new type of gas that is only used to price blob transactions. It is independent of normal gas and follows its own targeting rule. The new pricing rule requires the `excess_data_gas` header field. The actual `data_fee` as calculated via `calc_data_fee` (see below) is deducted from the sender balance before transaction execution and burned, and is not refunded in case of transaction failure.\n",
    "\n",
    "The data gasprice update rule approximates the formula\n",
    "\n",
    "`data_gasprice = MIN_DATA_GASPRICE * e**(excess_data_gas / DATA_GASPRICE_UPDATE_FRACTION)`,\n",
    "\n",
    "where `excess_data_gas` is the total “extra” amount of data gas that the chain has consumed relative to the “targeted” amount to date. Like EIP-1559, it’s a self-correcting formula: as the excess goes higher, the `data_gasprice` increases exponentially, reducing usage and eventually forcing the excess back down.\n",
    "\n",
    "The block-by-block behavior is roughly as follows. If block `N` consumes `X` data gas, then in block `N+1` `excess_data_gas` increases by `X - TARGET_DATA_GAS_PER_BLOCK`, and so the `data_gasprice` of block `N+1` increases by a factor of \n",
    "\n",
    "`e**((X - TARGET_DATA_GAS_PER_BLOCK) / DATA_GASPRICE_UPDATE_FRACTION)`. \n",
    "\n",
    "Hence, it has a similar effect to the existing EIP-1559, but is more “stable” in the sense that it responds in the same way to the same total usage regardless of how it’s distributed.\n",
    "\n",
    "The parameter `DATA_GASPRICE_UPDATE_FRACTION` controls the maximum rate of change of the blob gas price. It is chosen to target a maximum change rate of `e(TARGET_DATA_GAS_PER_BLOCK / DATA_GASPRICE_UPDATE_FRACTION) ≈ 1.125` per block."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`Transaction` class:\n",
    "\n",
    "- `max_fee_per_gas`: the max price that user is willing to pay.\n",
    "- `max_priority_fee_per_gas`: the tip to the block producers.\n",
    "- `gas_used`: gas resource required to execute transaction.\n",
    "\n",
    "`BlobTransaction(Transaction)` class:\n",
    "\n",
    "- `max_fee_per_data_gas`: the max data gasprice that user is willing to pay.\n",
    "- `blob_hashes`: number of blobs included in transaction.\n",
    "\n",
    "\n",
    "`generate_demand` function: generates demand for transactions. This is defined in each simulation instance as assumptions change.\n",
    "\n",
    "`update_price` function: calculate resource prices using EIP-1559 formula according do deviation vs sustainable _resource target_: `new_price = price * (1 + ((utilized - target) / target) / 8)`.\n",
    "\n",
    "`update_excess_data_gas` function: the excess data gas consumed (above the sustainable target) is computed according to the gas accounting formulas for blobs specified in the EIP.\n",
    "\n",
    "`build_block` function: selects transactions, steps (1) filter out invalid txns for which gahimit is not enough and (2) build block respecting _resource limits_. Greedy block building logic where builder sorts by total gas premium and then selects transactions up to resource limit."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import secrets\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from radcad import Model, Simulation, Experiment\n",
    "from radcad.engine import Engine, Backend\n",
    "from random import randint, uniform\n",
    "\n",
    "sns.set(style=\"whitegrid\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "constants = {\n",
    "    # 1559 fee mechanism params \n",
    "    \"BLOCK_RESOURCE_LIMITS\": 30e6,\n",
    "    \"BLOCK_RESOURCE_TARGETS\": 15e6,\n",
    "    \"BASEFEE_MAX_CHANGE_DENOMINATOR\": 8,\n",
    "    \n",
    "    # 4844 fee mechanism params\n",
    "    \"MAX_DATA_GAS_PER_BLOCK\": 2**19,\n",
    "    \"TARGET_DATA_GAS_PER_BLOCK\": 2**18,\n",
    "    \"DATA_GAS_PER_BLOB\": 2**17,\n",
    "    \"MIN_DATA_GASPRICE\": 1,\n",
    "    \"DATA_GASPRICE_UPDATE_FRACTION\": 2225652,\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# helper function approximating exponential with taylor\n",
    "def fake_exponential(factor: int, numerator: int, denominator: int) -> int:\n",
    "    i = 1\n",
    "    output = 0\n",
    "    numerator_accum = factor * denominator\n",
    "    while numerator_accum > 0:\n",
    "        output += numerator_accum\n",
    "        numerator_accum = (numerator_accum * numerator) // (denominator * i)\n",
    "        i += 1\n",
    "    return output // denominator\n",
    "\n",
    "def calc_price_data(excess_data_gas) -> int:\n",
    "    return fake_exponential(\n",
    "        constants[\"MIN_DATA_GASPRICE\"],\n",
    "        excess_data_gas,\n",
    "        constants[\"DATA_GASPRICE_UPDATE_FRACTION\"]\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Transaction:\n",
    "    def __init__(self, max_priority_fee_per_gas, gas_used, max_fee_per_gas):\n",
    "        self.tx_hash = secrets.token_bytes(6).hex()\n",
    "        self.max_fee_per_gas = max_fee_per_gas\n",
    "        self.max_priority_fee_per_gas = max_priority_fee_per_gas\n",
    "        self.gas_used = gas_used\n",
    "\n",
    "    def is_valid(self, price, price_data):\n",
    "        return self.max_fee_per_gas >= price\n",
    "    \n",
    "    def get_premium(self, price):\n",
    "        priority_fee = min(self.max_fee_per_gas - price, self.max_priority_fee_per_gas)\n",
    "        return self.gas_used * priority_fee\n",
    "\n",
    "class BlobTransaction(Transaction):\n",
    "    def __init__(self, max_priority_fee_per_gas, gas_used, max_fee_per_gas, max_fee_per_data_gas, blob_hashes):\n",
    "        super().__init__(max_priority_fee_per_gas, gas_used, max_fee_per_gas)\n",
    "        self.max_fee_per_data_gas = max_fee_per_data_gas\n",
    "        self.blob_hashes = blob_hashes\n",
    "    \n",
    "    def is_valid(self, price, price_data):\n",
    "        return self.max_fee_per_gas >= price and self.max_fee_per_data_gas >= price_data\n",
    "\n",
    "class Block():\n",
    "    def __init__(self, txs):\n",
    "        self.txs = txs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 1559 fee mechanism\n",
    "def update_price(params, step, h, s, _input):\n",
    "    block = _input[\"block\"]\n",
    "    price = s[\"price\"]\n",
    "    \n",
    "    target = constants[\"BLOCK_RESOURCE_TARGETS\"]\n",
    "    utilized = sum([tx.gas_used for tx in block.txs]) \n",
    "    new_price = price * (1 + (utilized - target) / (target * constants[\"BASEFEE_MAX_CHANGE_DENOMINATOR\"]))\n",
    "    \n",
    "    return (\"price\", new_price)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 4844 fee mechanism\n",
    "def update_excess_data_gas(params, step, h, s, _input):\n",
    "    block = _input[\"block\"]\n",
    "    excess_data_gas = s[\"excess_data_gas\"]\n",
    "    \n",
    "    target = constants[\"TARGET_DATA_GAS_PER_BLOCK\"]\n",
    "    utilized = constants[\"DATA_GAS_PER_BLOB\"] * sum([tx.blob_hashes if isinstance(tx, BlobTransaction) else 0 for tx in block.txs]) \n",
    "    new_excess_data_gas = max(excess_data_gas + (utilized - target), 0)\n",
    "\n",
    "    return (\"excess_data_gas\", new_excess_data_gas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def build_block(params, step, h, s):\n",
    "    demand = s[\"demand\"]\n",
    "    price = s[\"price\"]\n",
    "    excess_data_gas = s[\"excess_data_gas\"]\n",
    "    price_data = calc_price_data(excess_data_gas)\n",
    "    \n",
    "    # select valid transactions and sort them by total gas premium\n",
    "    # for BlobTransaction we check validity against price_data too\n",
    "    sorted_valid_demand = sorted(\n",
    "        [tx for tx_hash, tx in demand.items() if tx.is_valid(price, price_data)],\n",
    "        key = lambda tx: -tx.get_premium(price)\n",
    "    )\n",
    "    \n",
    "    gas_limit = constants[\"BLOCK_RESOURCE_LIMITS\"]\n",
    "    data_limit = constants[\"MAX_DATA_GAS_PER_BLOCK\"]\n",
    "    included_transactions = []\n",
    "    utilized = 0\n",
    "    utilized_data = 0\n",
    "    for tx in sorted_valid_demand:\n",
    "        if utilized<=gas_limit:\n",
    "            if isinstance(tx, BlobTransaction) and utilized_data<=data_limit:\n",
    "                included_transactions.append(tx)\n",
    "                utilized += tx.gas_used\n",
    "                utilized_data += tx.blob_hashes * constants[\"DATA_GAS_PER_BLOB\"]\n",
    "            elif isinstance(tx, BlobTransaction) and utilized_data>data_limit:\n",
    "                continue\n",
    "            else:\n",
    "                included_transactions.append(tx)\n",
    "                utilized += tx.gas_used\n",
    "        else:\n",
    "            break\n",
    "\n",
    "    return ({\"block\": Block(txs=included_transactions)})\n",
    "\n",
    "\n",
    "def record_latest_block(params, step, h, s, _input):\n",
    "    block = _input[\"block\"]\n",
    "    return (\"latest_block\", block)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Simulations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Basic Case"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_demand(params, step, h, s, _input):\n",
    "\n",
    "    demand = {}  \n",
    "    for i in range(1000):\n",
    "        tx = Transaction(\n",
    "            gas_used = 50000,\n",
    "            max_fee_per_gas = 100*uniform(0.75,1.25),\n",
    "            max_priority_fee_per_gas = uniform(1, 10), # ~1/10 of max fee to represent builder \"cost\"\n",
    "        )\n",
    "        demand[tx.tx_hash] = tx\n",
    "\n",
    "    for i in range(10):\n",
    "        tx = BlobTransaction(\n",
    "            gas_used = 50000,\n",
    "            max_fee_per_gas = 150*uniform(0.75,1.25),\n",
    "            max_priority_fee_per_gas = uniform(1, 10),\n",
    "            max_fee_per_data_gas = 50*uniform(0.75,1.25),\n",
    "            blob_hashes = 2\n",
    "        )\n",
    "        demand[tx.tx_hash] = tx\n",
    "\n",
    "    return (\"demand\", demand)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "\n",
    "psub = [{\n",
    "    \"policies\": {},\n",
    "    \"variables\": {\n",
    "        \"demand\": generate_demand\n",
    "    }\n",
    "}, {\n",
    "    \"policies\": {\n",
    "        \"action\": build_block\n",
    "    },\n",
    "    \"variables\": {\n",
    "        \"excess_data_gas\": update_excess_data_gas,\n",
    "        \"price\": update_price,\n",
    "        \"latest_block\": record_latest_block,\n",
    "    }\n",
    "}]\n",
    "\n",
    "initial_conditions = {\n",
    "    \"price\": 1,\n",
    "    \"excess_data_gas\": 0,\n",
    "    \"demand\": {},\n",
    "    \"latest_block\": Block(txs=[])\n",
    "}\n",
    "\n",
    "simulation_parameters = {\n",
    "    'T': 100,\n",
    "    'N': 1,\n",
    "    'M': {}\n",
    "}\n",
    "\n",
    "model = Model(\n",
    "    initial_state=initial_conditions,\n",
    "    state_update_blocks=psub,\n",
    "    params=simulation_parameters\n",
    ")\n",
    "\n",
    "simulation = Simulation(\n",
    "    model=model,\n",
    "    timesteps=simulation_parameters['T'],\n",
    "    runs=simulation_parameters['N']\n",
    ")\n",
    "\n",
    "experiment = Experiment(simulations=[simulation])\n",
    "experiment.engine=Engine(backend=Backend.PATHOS, drop_substeps=True)\n",
    "df = pd.DataFrame(experiment.run())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>excess_data_gas</th>\n",
       "      <th>demand</th>\n",
       "      <th>latest_block</th>\n",
       "      <th>simulation</th>\n",
       "      <th>subset</th>\n",
       "      <th>run</th>\n",
       "      <th>substep</th>\n",
       "      <th>timestep</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>{}</td>\n",
       "      <td>&lt;__main__.Block object at 0x12f301990&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.125417</td>\n",
       "      <td>524288</td>\n",
       "      <td>{'983b7a336f27': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x130b2e8c0&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.266563</td>\n",
       "      <td>1048576</td>\n",
       "      <td>{'876830ab2f19': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x12f422380&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.425411</td>\n",
       "      <td>1572864</td>\n",
       "      <td>{'dc4d63cfb716': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x130925e40&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.604181</td>\n",
       "      <td>2097152</td>\n",
       "      <td>{'620cc2c06cb7': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13021d900&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>109.096352</td>\n",
       "      <td>9175040</td>\n",
       "      <td>{'4e93e03c7a4d': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13365ec50&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>109.460007</td>\n",
       "      <td>8912896</td>\n",
       "      <td>{'38f4bddeaa16': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x133622710&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>110.737040</td>\n",
       "      <td>9437184</td>\n",
       "      <td>{'b403ddb8c45b': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x1331fa1d0&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>109.629670</td>\n",
       "      <td>9175040</td>\n",
       "      <td>{'488d44c0c35d': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x132b55630&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100</th>\n",
       "      <td>111.182757</td>\n",
       "      <td>8912896</td>\n",
       "      <td>{'d40ba9eb23ad': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x135fb1750&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>100</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>101 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          price  excess_data_gas  \\\n",
       "0      1.000000                0   \n",
       "1      1.125417           524288   \n",
       "2      1.266563          1048576   \n",
       "3      1.425411          1572864   \n",
       "4      1.604181          2097152   \n",
       "..          ...              ...   \n",
       "96   109.096352          9175040   \n",
       "97   109.460007          8912896   \n",
       "98   110.737040          9437184   \n",
       "99   109.629670          9175040   \n",
       "100  111.182757          8912896   \n",
       "\n",
       "                                                demand  \\\n",
       "0                                                   {}   \n",
       "1    {'983b7a336f27': <__main__.Transaction object ...   \n",
       "2    {'876830ab2f19': <__main__.Transaction object ...   \n",
       "3    {'dc4d63cfb716': <__main__.Transaction object ...   \n",
       "4    {'620cc2c06cb7': <__main__.Transaction object ...   \n",
       "..                                                 ...   \n",
       "96   {'4e93e03c7a4d': <__main__.Transaction object ...   \n",
       "97   {'38f4bddeaa16': <__main__.Transaction object ...   \n",
       "98   {'b403ddb8c45b': <__main__.Transaction object ...   \n",
       "99   {'488d44c0c35d': <__main__.Transaction object ...   \n",
       "100  {'d40ba9eb23ad': <__main__.Transaction object ...   \n",
       "\n",
       "                               latest_block  simulation  subset  run  substep  \\\n",
       "0    <__main__.Block object at 0x12f301990>           0       0    1        0   \n",
       "1    <__main__.Block object at 0x130b2e8c0>           0       0    1        2   \n",
       "2    <__main__.Block object at 0x12f422380>           0       0    1        2   \n",
       "3    <__main__.Block object at 0x130925e40>           0       0    1        2   \n",
       "4    <__main__.Block object at 0x13021d900>           0       0    1        2   \n",
       "..                                      ...         ...     ...  ...      ...   \n",
       "96   <__main__.Block object at 0x13365ec50>           0       0    1        2   \n",
       "97   <__main__.Block object at 0x133622710>           0       0    1        2   \n",
       "98   <__main__.Block object at 0x1331fa1d0>           0       0    1        2   \n",
       "99   <__main__.Block object at 0x132b55630>           0       0    1        2   \n",
       "100  <__main__.Block object at 0x135fb1750>           0       0    1        2   \n",
       "\n",
       "     timestep  \n",
       "0           0  \n",
       "1           1  \n",
       "2           2  \n",
       "3           3  \n",
       "4           4  \n",
       "..        ...  \n",
       "96         96  \n",
       "97         97  \n",
       "98         98  \n",
       "99         99  \n",
       "100       100  \n",
       "\n",
       "[101 rows x 9 columns]"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='timestep'>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[df.substep == 2].plot('timestep', ['price'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='timestep'>"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[df.substep == 2].plot('timestep', ['excess_data_gas'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='timestep'>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['price_data'] = df['excess_data_gas'].apply(lambda x: calc_price_data(x))\n",
    "df[df.substep == 2].plot('timestep', ['price', 'price_data'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Backtest switching to blob txs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now do a backtest considering the 98%+ percent of the current batch-data load that is given by Arbitrum and Optimism. We chose the day with the highest batch-data load in the first two months of 2023, February 24th.\n",
    "\n",
    "- Arbitrum posted 1055 consistent batches of ~99kb (600b stdev) every 6.78 blocks on avg, for a total of ~100mb/day\n",
    "- Optimism posted 2981 variable batches of avg ~31kb (19kb stdev) every 2.37 blocks on avg, for a total of ~93mb/day\n",
    "\n",
    "Note that all gas accounting in 4844 is done using blobs and their fixed capacity size of 2**17b (~125kb).\n",
    "\n",
    "**Case 1** considers a scenario with a data load which is slightly higher (1.5x) than the historical data from the peak day considered. For convenience, we assume that L2s do not optimize their batch sizes (okay for ARB but suboptimal for OP) which results in a likely overestimation of the blobs demand. Results show that this load is not enough to start moving the data price.\n",
    "\n",
    "**Case 2** consider the case in which data load is inflated by roughly 10x, a value that is just enough to kickstart the data price discovery process.\n",
    "\n",
    "We start with loading historical data (for original query and more stats for this data check out the data-processing-utils notebook in the same folder)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>block_number</th>\n",
       "      <th>index</th>\n",
       "      <th>gas_used</th>\n",
       "      <th>max_fee</th>\n",
       "      <th>max_priority_fee</th>\n",
       "      <th>base_fee</th>\n",
       "      <th>data_tag</th>\n",
       "      <th>data_bytes</th>\n",
       "      <th>gas_used_calldata</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>16694584</td>\n",
       "      <td>14</td>\n",
       "      <td>104935</td>\n",
       "      <td>4.595416e+10</td>\n",
       "      <td>5.000000e+08</td>\n",
       "      <td>32301451293</td>\n",
       "      <td>other</td>\n",
       "      <td>132.0</td>\n",
       "      <td>2112.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>16694584</td>\n",
       "      <td>57</td>\n",
       "      <td>119600</td>\n",
       "      <td>4.140369e+10</td>\n",
       "      <td>1.000000e+09</td>\n",
       "      <td>32301451293</td>\n",
       "      <td>other</td>\n",
       "      <td>36.0</td>\n",
       "      <td>576.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>16694584</td>\n",
       "      <td>36</td>\n",
       "      <td>110595</td>\n",
       "      <td>4.600000e+10</td>\n",
       "      <td>2.000000e+09</td>\n",
       "      <td>32301451293</td>\n",
       "      <td>other</td>\n",
       "      <td>36.0</td>\n",
       "      <td>576.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>16694584</td>\n",
       "      <td>106</td>\n",
       "      <td>110595</td>\n",
       "      <td>4.595416e+10</td>\n",
       "      <td>5.000000e+08</td>\n",
       "      <td>32301451293</td>\n",
       "      <td>other</td>\n",
       "      <td>36.0</td>\n",
       "      <td>576.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>16694584</td>\n",
       "      <td>47</td>\n",
       "      <td>105818</td>\n",
       "      <td>6.848287e+10</td>\n",
       "      <td>1.109231e+09</td>\n",
       "      <td>32301451293</td>\n",
       "      <td>other</td>\n",
       "      <td>164.0</td>\n",
       "      <td>2624.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   block_number  index  gas_used       max_fee  max_priority_fee     base_fee  \\\n",
       "0      16694584     14    104935  4.595416e+10      5.000000e+08  32301451293   \n",
       "1      16694584     57    119600  4.140369e+10      1.000000e+09  32301451293   \n",
       "2      16694584     36    110595  4.600000e+10      2.000000e+09  32301451293   \n",
       "3      16694584    106    110595  4.595416e+10      5.000000e+08  32301451293   \n",
       "4      16694584     47    105818  6.848287e+10      1.109231e+09  32301451293   \n",
       "\n",
       "  data_tag  data_bytes  gas_used_calldata  \n",
       "0    other       132.0             2112.0  \n",
       "1    other        36.0              576.0  \n",
       "2    other        36.0              576.0  \n",
       "3    other        36.0              576.0  \n",
       "4    other       164.0             2624.0  "
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "demand_input = pd.read_csv('4844-sample-new.csv', index_col=False)\n",
    "demand_input.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Case 1: historical demand x1.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_demand_from_data(params, step, h, s, _input):\n",
    "\n",
    "    di = demand_input[demand_input['block_number']==(min(demand_input['block_number'])+len(h))].to_dict('index')\n",
    "\n",
    "    demand = {}\n",
    "    for key, val in di.items():\n",
    "        if val['data_tag']=='other':\n",
    "            tx = Transaction(\n",
    "                gas_used = val['gas_used'],\n",
    "                max_fee_per_gas = val['max_fee'],\n",
    "                max_priority_fee_per_gas = val['max_priority_fee'],\n",
    "            )\n",
    "        else:\n",
    "            tx = BlobTransaction(\n",
    "                gas_used = 50000,\n",
    "                max_fee_per_gas = val['max_fee'],\n",
    "                max_priority_fee_per_gas = val['max_priority_fee'],\n",
    "                max_fee_per_data_gas = 50e9,\n",
    "                blob_hashes = 1\n",
    "            )\n",
    "        demand[tx.tx_hash] = tx\n",
    "\n",
    "    return (\"demand\", demand)\n",
    "\n",
    "\n",
    "def build_block_from_data(params, step, h, s):\n",
    "    demand = s[\"demand\"]\n",
    "    price = s[\"price\"]\n",
    "    excess_data_gas = s[\"excess_data_gas\"]\n",
    "    price_data = calc_price_data(excess_data_gas)\n",
    "    \n",
    "    # REMOVE selection, include everything for now\n",
    "    sorted_valid_demand = [tx for tx_hash, tx in demand.items()]\n",
    "    \n",
    "    gas_limit = constants[\"BLOCK_RESOURCE_LIMITS\"]\n",
    "    data_limit = constants[\"MAX_DATA_GAS_PER_BLOCK\"]\n",
    "    included_transactions = []\n",
    "    utilized = 0\n",
    "    utilized_data = 0\n",
    "    for tx in sorted_valid_demand:\n",
    "        if utilized<=gas_limit:\n",
    "            if isinstance(tx, BlobTransaction) and utilized_data<=data_limit:\n",
    "                included_transactions.append(tx)\n",
    "                utilized += tx.gas_used\n",
    "                utilized_data += tx.blob_hashes * constants[\"DATA_GAS_PER_BLOB\"]\n",
    "            elif isinstance(tx, BlobTransaction) and utilized_data>data_limit:\n",
    "                continue\n",
    "            else:\n",
    "                included_transactions.append(tx)\n",
    "                utilized += tx.gas_used\n",
    "        else:\n",
    "            break\n",
    "\n",
    "    return ({\"block\": Block(txs=included_transactions)})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "\n",
    "psub = [{\n",
    "    \"policies\": {},\n",
    "    \"variables\": {\n",
    "        \"demand\": generate_demand_from_data\n",
    "    }\n",
    "}, {\n",
    "    \"policies\": {\n",
    "        \"action\": build_block_from_data\n",
    "    },\n",
    "    \"variables\": {\n",
    "        \"excess_data_gas\": update_excess_data_gas,\n",
    "        \"price\": update_price,\n",
    "        \"latest_block\": record_latest_block,\n",
    "    }\n",
    "}]\n",
    "\n",
    "initial_conditions = {\n",
    "    \"price\": 28543674071,\n",
    "    \"excess_data_gas\": 0,\n",
    "    \"demand\": {},\n",
    "    \"latest_block\": Block(txs=[])\n",
    "}\n",
    "\n",
    "simulation_parameters = {\n",
    "    'T': 1600,\n",
    "    'N': 1,\n",
    "    'M': {}\n",
    "}\n",
    "\n",
    "model = Model(\n",
    "    initial_state=initial_conditions,\n",
    "    state_update_blocks=psub,\n",
    "    params=simulation_parameters\n",
    ")\n",
    "\n",
    "simulation = Simulation(\n",
    "    model=model,\n",
    "    timesteps=simulation_parameters['T'],\n",
    "    runs=simulation_parameters['N']\n",
    ")\n",
    "\n",
    "experiment = Experiment(simulations=[simulation])\n",
    "experiment.engine=Engine(backend=Backend.PATHOS, drop_substeps=True)\n",
    "df = pd.DataFrame(experiment.run())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>excess_data_gas</th>\n",
       "      <th>demand</th>\n",
       "      <th>latest_block</th>\n",
       "      <th>simulation</th>\n",
       "      <th>subset</th>\n",
       "      <th>run</th>\n",
       "      <th>substep</th>\n",
       "      <th>timestep</th>\n",
       "      <th>price_data</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>391</th>\n",
       "      <td>1.084784e+10</td>\n",
       "      <td>0</td>\n",
       "      <td>{b'\\\\\\xb0\\xdb\\x85\\x12\\xed\\x90D': &lt;__main__.Tra...</td>\n",
       "      <td>&lt;__main__.Block object at 0x138861f00&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>391</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>392</th>\n",
       "      <td>1.089326e+10</td>\n",
       "      <td>0</td>\n",
       "      <td>{b'\\xd8\\xd2,\\x92 \\x04\\xcb)': &lt;__main__.Transac...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13886e530&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>392</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>393</th>\n",
       "      <td>1.044866e+10</td>\n",
       "      <td>0</td>\n",
       "      <td>{b'\\xfbZJ\\xdd\\xdaO\\xea\\xe7': &lt;__main__.Transac...</td>\n",
       "      <td>&lt;__main__.Block object at 0x138881d80&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>393</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>394</th>\n",
       "      <td>1.026659e+10</td>\n",
       "      <td>0</td>\n",
       "      <td>{b'\\x8aP\\x18\\xa1\\x1f\\x8f\\xfe\\x08': &lt;__main__.T...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13888e830&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>394</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>395</th>\n",
       "      <td>1.066654e+10</td>\n",
       "      <td>0</td>\n",
       "      <td>{b'\\xb7\\xb3C\\x11=O\\xad#': &lt;__main__.Transactio...</td>\n",
       "      <td>&lt;__main__.Block object at 0x1388ac1f0&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>395</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>396</th>\n",
       "      <td>1.093064e+10</td>\n",
       "      <td>262144</td>\n",
       "      <td>{b'\\xba\\xcb\\xa7\\xacs\\xea\\xe8i': &lt;__main__.Tran...</td>\n",
       "      <td>&lt;__main__.Block object at 0x1388b9450&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>396</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>397</th>\n",
       "      <td>1.051280e+10</td>\n",
       "      <td>0</td>\n",
       "      <td>{b'\\xb5\\t\\x18\\xce\\x187\\x1d\\x10': &lt;__main__.Tra...</td>\n",
       "      <td>&lt;__main__.Block object at 0x1388d1b40&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>397</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>398</th>\n",
       "      <td>1.034515e+10</td>\n",
       "      <td>0</td>\n",
       "      <td>{b'\\x85+8\\x8aU0j\\x19': &lt;__main__.Transaction o...</td>\n",
       "      <td>&lt;__main__.Block object at 0x1388e28c0&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>398</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>399</th>\n",
       "      <td>9.331096e+09</td>\n",
       "      <td>0</td>\n",
       "      <td>{b'\\x8aN\\xd6\\xad\\x98\\xbc\\xac\\xe1': &lt;__main__.T...</td>\n",
       "      <td>&lt;__main__.Block object at 0x1388e3e80&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>399</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>400</th>\n",
       "      <td>1.013450e+10</td>\n",
       "      <td>0</td>\n",
       "      <td>{b'\\xf9B.\\x84\\xa4\\x0c)\\t': &lt;__main__.Transacti...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13890d990&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>400</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            price  excess_data_gas  \\\n",
       "391  1.084784e+10                0   \n",
       "392  1.089326e+10                0   \n",
       "393  1.044866e+10                0   \n",
       "394  1.026659e+10                0   \n",
       "395  1.066654e+10                0   \n",
       "396  1.093064e+10           262144   \n",
       "397  1.051280e+10                0   \n",
       "398  1.034515e+10                0   \n",
       "399  9.331096e+09                0   \n",
       "400  1.013450e+10                0   \n",
       "\n",
       "                                                demand  \\\n",
       "391  {b'\\\\\\xb0\\xdb\\x85\\x12\\xed\\x90D': <__main__.Tra...   \n",
       "392  {b'\\xd8\\xd2,\\x92 \\x04\\xcb)': <__main__.Transac...   \n",
       "393  {b'\\xfbZJ\\xdd\\xdaO\\xea\\xe7': <__main__.Transac...   \n",
       "394  {b'\\x8aP\\x18\\xa1\\x1f\\x8f\\xfe\\x08': <__main__.T...   \n",
       "395  {b'\\xb7\\xb3C\\x11=O\\xad#': <__main__.Transactio...   \n",
       "396  {b'\\xba\\xcb\\xa7\\xacs\\xea\\xe8i': <__main__.Tran...   \n",
       "397  {b'\\xb5\\t\\x18\\xce\\x187\\x1d\\x10': <__main__.Tra...   \n",
       "398  {b'\\x85+8\\x8aU0j\\x19': <__main__.Transaction o...   \n",
       "399  {b'\\x8aN\\xd6\\xad\\x98\\xbc\\xac\\xe1': <__main__.T...   \n",
       "400  {b'\\xf9B.\\x84\\xa4\\x0c)\\t': <__main__.Transacti...   \n",
       "\n",
       "                               latest_block  simulation  subset  run  substep  \\\n",
       "391  <__main__.Block object at 0x138861f00>           0       0    1        2   \n",
       "392  <__main__.Block object at 0x13886e530>           0       0    1        2   \n",
       "393  <__main__.Block object at 0x138881d80>           0       0    1        2   \n",
       "394  <__main__.Block object at 0x13888e830>           0       0    1        2   \n",
       "395  <__main__.Block object at 0x1388ac1f0>           0       0    1        2   \n",
       "396  <__main__.Block object at 0x1388b9450>           0       0    1        2   \n",
       "397  <__main__.Block object at 0x1388d1b40>           0       0    1        2   \n",
       "398  <__main__.Block object at 0x1388e28c0>           0       0    1        2   \n",
       "399  <__main__.Block object at 0x1388e3e80>           0       0    1        2   \n",
       "400  <__main__.Block object at 0x13890d990>           0       0    1        2   \n",
       "\n",
       "     timestep  price_data  \n",
       "391       391           1  \n",
       "392       392           1  \n",
       "393       393           1  \n",
       "394       394           1  \n",
       "395       395           1  \n",
       "396       396           1  \n",
       "397       397           1  \n",
       "398       398           1  \n",
       "399       399           1  \n",
       "400       400           1  "
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['price'].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='timestep'>"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['price_data'] = df['excess_data_gas'].apply(lambda x: calc_price_data(x))\n",
    "df[df.substep == 2].plot('timestep', ['price_data'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='block_number'>"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "demand_input[['block_number', 'base_fee']].groupby('block_number').min().head(1600).plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Additional note: Comparing the gas price to the base fee from the original data we see that moving to batches to blobs as a deflationary effect on normal gas price (as expected)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Case 2: historical demand x3 + 4 blobs every 3 blocks = historical demand ~x10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_demand_from_data(params, step, h, s, _input):\n",
    "\n",
    "    di = demand_input[demand_input['block_number']==(min(demand_input['block_number'])+len(h))].to_dict('index')\n",
    "    \n",
    "    demand = {}\n",
    "    for key, val in di.items():\n",
    "        if val['data_tag']=='other':\n",
    "            tx = Transaction(\n",
    "                gas_used = val['gas_used'],\n",
    "                max_fee_per_gas = val['max_fee'],\n",
    "                max_priority_fee_per_gas = val['max_priority_fee'],\n",
    "            )\n",
    "        else:\n",
    "            tx = BlobTransaction(\n",
    "                gas_used = 50000,\n",
    "                max_fee_per_gas = val['max_fee'],\n",
    "                max_priority_fee_per_gas = val['max_priority_fee'],\n",
    "                max_fee_per_data_gas = 50e9*uniform(0.75,1.25),\n",
    "                blob_hashes = 2\n",
    "            )\n",
    "        demand[tx.tx_hash] = tx\n",
    "    \n",
    "    if (len(h) % 3)==0:\n",
    "        for i in range(2):\n",
    "            tx = BlobTransaction(\n",
    "                gas_used = 50000,\n",
    "                max_fee_per_gas = 150e9*uniform(0.75,1.25),\n",
    "                max_priority_fee_per_gas = 1e9*uniform(1, 10),\n",
    "                max_fee_per_data_gas = 50e9*uniform(0.75,1.25),\n",
    "                blob_hashes = 2\n",
    "            )\n",
    "            demand[tx.tx_hash] = tx\n",
    "\n",
    "    return (\"demand\", demand)\n",
    "\n",
    "\n",
    "def build_block_from_data(params, step, h, s):\n",
    "    demand = s[\"demand\"]\n",
    "    price = s[\"price\"]\n",
    "    excess_data_gas = s[\"excess_data_gas\"]\n",
    "    price_data = calc_price_data(excess_data_gas)\n",
    "\n",
    "    # select valid transactions and sort them by total gas premium\n",
    "    # for BlobTransaction we check validity against price_data too\n",
    "    sorted_valid_demand = sorted(\n",
    "        [tx for tx_hash, tx in demand.items() if tx.is_valid(price, price_data)],\n",
    "        key = lambda tx: -tx.get_premium(price)\n",
    "    )\n",
    "    \n",
    "    gas_limit = constants[\"BLOCK_RESOURCE_LIMITS\"]\n",
    "    data_limit = constants[\"MAX_DATA_GAS_PER_BLOCK\"]\n",
    "    included_transactions = []\n",
    "    utilized = 0\n",
    "    utilized_data = 0\n",
    "    for tx in sorted_valid_demand:\n",
    "        if utilized<=gas_limit:\n",
    "            if isinstance(tx, BlobTransaction) and utilized_data<=data_limit:\n",
    "                included_transactions.append(tx)\n",
    "                utilized += tx.gas_used\n",
    "                utilized_data += tx.blob_hashes * constants[\"DATA_GAS_PER_BLOB\"]\n",
    "            elif isinstance(tx, BlobTransaction) and utilized_data>data_limit:\n",
    "                continue\n",
    "            else:\n",
    "                included_transactions.append(tx)\n",
    "                utilized += tx.gas_used\n",
    "        else:\n",
    "            break\n",
    "\n",
    "    return ({\"block\": Block(txs=included_transactions)})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "%%capture\n",
    "\n",
    "psub = [{\n",
    "    \"policies\": {},\n",
    "    \"variables\": {\n",
    "        \"demand\": generate_demand_from_data\n",
    "    }\n",
    "}, {\n",
    "    \"policies\": {\n",
    "        \"action\": build_block_from_data\n",
    "    },\n",
    "    \"variables\": {\n",
    "        \"excess_data_gas\": update_excess_data_gas,\n",
    "        \"price\": update_price,\n",
    "        \"latest_block\": record_latest_block,\n",
    "    }\n",
    "}]\n",
    "\n",
    "initial_conditions = {\n",
    "    \"price\": 28543674071,\n",
    "    \"excess_data_gas\": 0,\n",
    "    \"demand\": {},\n",
    "    \"latest_block\": Block(txs=[])\n",
    "}\n",
    "\n",
    "simulation_parameters = {\n",
    "    'T': 1600,\n",
    "    'N': 1,\n",
    "    'M': {}\n",
    "}\n",
    "\n",
    "model = Model(\n",
    "    initial_state=initial_conditions,\n",
    "    state_update_blocks=psub,\n",
    "    params=simulation_parameters\n",
    ")\n",
    "\n",
    "simulation = Simulation(\n",
    "    model=model,\n",
    "    timesteps=simulation_parameters['T'],\n",
    "    runs=simulation_parameters['N']\n",
    ")\n",
    "\n",
    "experiment = Experiment(simulations=[simulation])\n",
    "experiment.engine=Engine(backend=Backend.PATHOS, drop_substeps=True)\n",
    "df = pd.DataFrame(experiment.run())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "      <th>excess_data_gas</th>\n",
       "      <th>demand</th>\n",
       "      <th>latest_block</th>\n",
       "      <th>simulation</th>\n",
       "      <th>subset</th>\n",
       "      <th>run</th>\n",
       "      <th>substep</th>\n",
       "      <th>timestep</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1591</th>\n",
       "      <td>1.799472e+08</td>\n",
       "      <td>54525952</td>\n",
       "      <td>{'9b31375c41e0': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13ff58130&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1592</th>\n",
       "      <td>1.719275e+08</td>\n",
       "      <td>54263808</td>\n",
       "      <td>{'77bfa75cabbc': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13ff5ab30&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1592</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1593</th>\n",
       "      <td>1.832515e+08</td>\n",
       "      <td>54788096</td>\n",
       "      <td>{'a90e2d8fcb83': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13ff73490&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1593</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1594</th>\n",
       "      <td>1.780317e+08</td>\n",
       "      <td>54525952</td>\n",
       "      <td>{'6b808f278e10': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13ff8a170&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1594</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1595</th>\n",
       "      <td>1.703862e+08</td>\n",
       "      <td>54263808</td>\n",
       "      <td>{'1d1bd0806647': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13ffa0c40&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1596</th>\n",
       "      <td>1.672521e+08</td>\n",
       "      <td>54525952</td>\n",
       "      <td>{'e6eaf6fc8b8b': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13ffb83a0&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1596</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1597</th>\n",
       "      <td>1.657340e+08</td>\n",
       "      <td>54525952</td>\n",
       "      <td>{'5bd93244031b': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13ffbbcd0&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1597</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1598</th>\n",
       "      <td>1.647675e+08</td>\n",
       "      <td>54263808</td>\n",
       "      <td>{'a9ff4312bd5d': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13ffd3910&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1598</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1599</th>\n",
       "      <td>1.619081e+08</td>\n",
       "      <td>54788096</td>\n",
       "      <td>{'ab7f2de6370c': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x13ffebb80&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1600</th>\n",
       "      <td>1.611968e+08</td>\n",
       "      <td>54525952</td>\n",
       "      <td>{'34bb61e7695a': &lt;__main__.Transaction object ...</td>\n",
       "      <td>&lt;__main__.Block object at 0x140002980&gt;</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>1600</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             price  excess_data_gas  \\\n",
       "1591  1.799472e+08         54525952   \n",
       "1592  1.719275e+08         54263808   \n",
       "1593  1.832515e+08         54788096   \n",
       "1594  1.780317e+08         54525952   \n",
       "1595  1.703862e+08         54263808   \n",
       "1596  1.672521e+08         54525952   \n",
       "1597  1.657340e+08         54525952   \n",
       "1598  1.647675e+08         54263808   \n",
       "1599  1.619081e+08         54788096   \n",
       "1600  1.611968e+08         54525952   \n",
       "\n",
       "                                                 demand  \\\n",
       "1591  {'9b31375c41e0': <__main__.Transaction object ...   \n",
       "1592  {'77bfa75cabbc': <__main__.Transaction object ...   \n",
       "1593  {'a90e2d8fcb83': <__main__.Transaction object ...   \n",
       "1594  {'6b808f278e10': <__main__.Transaction object ...   \n",
       "1595  {'1d1bd0806647': <__main__.Transaction object ...   \n",
       "1596  {'e6eaf6fc8b8b': <__main__.Transaction object ...   \n",
       "1597  {'5bd93244031b': <__main__.Transaction object ...   \n",
       "1598  {'a9ff4312bd5d': <__main__.Transaction object ...   \n",
       "1599  {'ab7f2de6370c': <__main__.Transaction object ...   \n",
       "1600  {'34bb61e7695a': <__main__.Transaction object ...   \n",
       "\n",
       "                                latest_block  simulation  subset  run  \\\n",
       "1591  <__main__.Block object at 0x13ff58130>           0       0    1   \n",
       "1592  <__main__.Block object at 0x13ff5ab30>           0       0    1   \n",
       "1593  <__main__.Block object at 0x13ff73490>           0       0    1   \n",
       "1594  <__main__.Block object at 0x13ff8a170>           0       0    1   \n",
       "1595  <__main__.Block object at 0x13ffa0c40>           0       0    1   \n",
       "1596  <__main__.Block object at 0x13ffb83a0>           0       0    1   \n",
       "1597  <__main__.Block object at 0x13ffbbcd0>           0       0    1   \n",
       "1598  <__main__.Block object at 0x13ffd3910>           0       0    1   \n",
       "1599  <__main__.Block object at 0x13ffebb80>           0       0    1   \n",
       "1600  <__main__.Block object at 0x140002980>           0       0    1   \n",
       "\n",
       "      substep  timestep  \n",
       "1591        2      1591  \n",
       "1592        2      1592  \n",
       "1593        2      1593  \n",
       "1594        2      1594  \n",
       "1595        2      1595  \n",
       "1596        2      1596  \n",
       "1597        2      1597  \n",
       "1598        2      1598  \n",
       "1599        2      1599  \n",
       "1600        2      1600  "
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='timestep'>"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[df.substep == 2].plot('timestep', ['price'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='timestep'>"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df['price_data'] = df['excess_data_gas'].apply(lambda x: calc_price_data(x))\n",
    "df[df.substep == 2].plot('timestep', ['price_data'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Resources\n",
    "\n",
    "- [Dune query for explorations](https://dune.com/queries/2173699?sidebar=query-explorer)\n",
    "- [Dan Boneh lecture on EVM resources](https://cs251.stanford.edu/lectures/lecture8.pdf)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}