7.9 KiB
Ethereum 2.0 Phase 1 -- Shard Data Chains
tags: spec
, eth2.0
, casper
, sharding
NOTICE: This document is a work-in-progress for researchers and implementers. It reflects recent spec changes and takes precedence over the Python proof-of-concept implementation.
Introduction
This document represents the specification for Phase 1 of Ethereum 2.0 -- Shard Data Chains. Phase 1 depends on the implementation of Phase 0 -- The Beacon Chain.
Ethereum 2.0 consists of a central beacon chain along with SHARD_COUNT
shard chains. Phase 1 is primarily concerned with the construction, validity, and consensus on the data of these shard chains. Phase 1 does not specify shard chain state execution or account balances. This is left for future phases.
Terminology
Constants
Phase 1 depends upon all of the constants defined in Phase 0 in addition to the following:
Constant | Value | Unit | Approximation |
---|---|---|---|
CHUNK_SIZE |
2**8 (= 256) | bytes | |
MAX_SHARD_BLOCK_SIZE |
2**15 (= 32768) | bytes |
Data Structures
Shard chain blocks
A ShardBlock
object has the following fields:
{
# Slot number
'slot': 'uint64',
# What shard is it on
'shard_id': 'uint64',
# Parent block hash
'parent_hash': 'hash32',
# Beacon chain block
'beacon_chain_ref': 'hash32',
# Depth of the Merkle tree
'data_tree_depth': 'uint8',
# Merkle root of data
'data_root': 'hash32'
# State root (placeholder for now)
'state_root': 'hash32',
# Attestation (including block signature)
'attester_bitfield': 'bytes',
'aggregate_sig': ['uint256'],
}
Shard block processing
For a block on a shard to be processed by a node, the following conditions must be met:
- The
ShardBlock
pointed to byparent_hash
has already been processed and accepted - The signature for the block from the proposer (see below for definition) of that block is included along with the block in the network message object
To validate a block header on shard shard_id
, compute as follows:
- Verify that
beacon_chain_ref
is the hash of a block in the beacon chain with slot less than or equal toslot
. Verify thatbeacon_chain_ref
is equal to or a descendant of thebeacon_chain_ref
specified in theShardBlock
pointed to byparent_hash
. - Let
state
be the state of the beacon chain block referred to bybeacon_chain_ref
. Letvalidators
be[validators[i] for i in state.current_persistent_committees[shard_id]]
. - Assert
len(attester_bitfield) == ceil_div8(len(validators))
- Let
curblock_proposer_index = hash(state.randao_mix + bytes8(shard_id) + bytes8(slot)) % len(validators)
. Letparent_proposer_index
be the same value calculated for the parent block. - Make sure that the
parent_proposer_index
'th bit in theattester_bitfield
is set to 1. - Generate the group public key by adding the public keys of all the validators for whom the corresponding position in the bitfield is set to 1. Verify the
aggregate_sig
using this as the pubkey and theparent_hash
as the message.
Verifying shard block data
At network layer, we expect a shard block header to be broadcast along with its block_body
. First, we define a helper function that takes as input beacon chain state and outputs the max block size in bytes:
def shard_block_maxbytes(state):
max_grains = MAX_SHARD_BLOCK_SIZE // CHUNK_SIZE
validators_at_target_committee_size = SHARD_COUNT * TARGET_COMMITTEE_SIZE
# number of grains per block is proportional to the number of validators
# up until `validators_at_target_committee_size`
grains = min(
len(get_active_validator_indices(state.validators)) * max_grains // validators_at_target_committee_size,
max_grains
)
return CHUNK_SIZE * grains
- Verify that
len(block_body) == shard_block_maxbytes(state)
- Define
filler_bytes = next_power_of_2(len(block_body)) - len(block_body)
. Compute a simple binary Merkle tree ofblock_body + bytes([0] * filler_bytes)
and verify that the root equals thedata_root
in the header.
Verifying a crosslink
A node should sign a crosslink only if the following conditions hold. If a node has the capability to perform the required level of verification, it should NOT follow chains on which a crosslink for which these conditions do NOT hold has been included, or a sufficient number of signatures have been included that during the next state recalculation, a crosslink will be registered.
First, the conditions must recursively apply to the crosslink referenced in last_crosslink_hash
for the same shard (unless last_crosslink_hash
equals zero, in which case we are at the genesis).
Second, we verify the shard_block_combined_data_root
. Let h
be the slot immediately after the slot of the shard block included by the last crosslink, and h+n-1
be the slot number of the block directly referenced by the current shard_block_hash
. Let B[i]
be the block at slot h+i
in the shard chain. Let bodies[0] .... bodies[n-1]
be the bodies of these blocks and roots[0] ... roots[n-1]
the data roots. If there is a missing slot in the shard chain at position h+i
, then bodies[i] == b'\x00' * shard_block_maxbytes(state[i])
and roots[i]
be the Merkle root of the empty data. Define compute_merkle_root
be a simple Merkle root calculating function that takes as input a list of objects, where the list's length must be an exact power of two. Let state[i]
be the beacon chain state at height h+i
(if the beacon chain is missing a block at some slot, the state is unchanged), and depths[i]
be equal to log2(next_power_of_2(shard_block_maxbytes(state[i]) // CHUNK_SIZE))
(ie. the expected depth of the i'th data tree). We define the function for computing the combined data root as follows:
def get_zeroroot_at_depth(n):
o = b'\x00' * CHUNK_SIZE
for i in range(n):
o = hash(o + o)
return o
def mk_combined_data_root(depths, roots):
default_value = get_zeroroot_at_depth(max(depths))
data = [default_value for _ in range(next_power_of_2(len(roots)))]
for i, (depth, root) in enumerate(zip(depths, roots)):
value = root
for j in range(depth, max(depths)):
value = hash(value, get_zeroroot_at_depth(depth + j))
data[i] = value
return compute_merkle_root(data)
This outputs the root of a tree of the data roots, with the data roots all adjusted to have the same height if needed. The tree can also be viewed as a tree of all of the underlying data concatenated together, appropriately padded. Here is an equivalent definition that uses bodies instead of roots [TODO: check equivalence]:
def mk_combined_data_root(depths, bodies):
default_value = get_zeroroot_at_depth(max(depths))
padded_body_length = max([CHUNK_SIZE * 2**d for d in depths])
data = b''
for body in bodies:
padded_body = body + bytes([0] * (padded_body_length - len(body)))
data += padded_body
data += bytes([0] * (next_power_of_2(len(data)) - len(data))
return compute_merkle_root([data[pos:pos+CHUNK_SIZE] for pos in range(0, len(data), CHUNK_SIZE)])
Verify that the shard_block_combined_data_root
is the output of these functions.
Shard block fork choice rule
The fork choice rule for any shard is LMD GHOST using the validators currently assigned to that shard, but instead of being rooted in the genesis it is rooted in the block referenced in the most recent accepted crosslink (ie. state.crosslinks[shard].shard_block_hash
). Only blocks whose beacon_chain_ref
is the block in the main beacon chain at the specified slot
should be considered (if the beacon chain skips a slot, then the block at that slot is considered to be the block in the beacon chain at the highest slot lower than a slot).