The radius sort performance improvement in content lookups was
not implemented in the trace version.
Also cleanup some part of the logging related to uTP connection
setup.
Each branch node may have up to 16 sub-items - currently, these are
given VertexID based when they are first needed leading to a
mostly-random order of vertexid for each subitem.
Here, we pre-allocate all 16 vertex ids such that when a branch subitem
is filled, it already has a vertexid waiting for it. This brings several
important benefits:
* subitems are sorted and "close" in their id sequencing - this means
that when rocksdb stores them, they are likely to end up in the same
data block thus improving read efficiency
* because the ids are consequtive, we can store just the starting id and
a bitmap representing which subitems are in use - this reduces disk
space usage for branches allowing more of them fit into a single disk
read, further improving disk read and caching performance - disk usage
at block 18M is down from 84 to 78gb!
* the in-memory footprint of VertexRef reduced allowing more instances
to fit into caches and less memory to be used overall.
Because of the increased locality of reference, it turns out that we no
longer need to iterate over the entire database to efficiently generate
the hash key database because the normal computation is now faster -
this significantly benefits "live" chain processing as well where each
dirtied key must be accompanied by a read of all branch subitems next to
it - most of the performance benefit in this branch comes from this
locality-of-reference improvement.
On a sample resync, there's already ~20% improvement with later blocks
seeing increasing benefit (because the trie is deeper in later blocks
leading to more benefit from branch read perf improvements)
```
blocks: 18729664, baseline: 190h43m49s, contender: 153h59m0s
Time (total): -36h44m48s, -19.27%
```
Note: clients need to be resynced as the PR changes the on-disk format
R.I.P. little bloom filter - your life in the repo was short but
valuable
* Kludge: fix `eip4844` import in `validate`
why:
Importing `validate` needs `blscurve` here or with the importing module.
* Separate out `FC` descriptor iinto separate file
why:
Needed for external descriptor access (e.g. for debugging)
* Debugging toolkit for `FC`
* Verify chain descriptor after changing state
The idea of the beacon-lc-bridge was to allow to bridge data into
the Portal network while only using p2p protocols to get access
to the data.
It is however incomplete as for history content the receipts are
missing. These could be added by also adding devp2p access.
But for the beacon content, there would be no way for getting the
historical summaries over p2p.
And then we did not even look yet on how to do this for state.
Considering it is incomplete it was also not being used by anyone
and thus we remove it.
There is an assertion hitting due to the additon of an iterator
that deletes items from the sequence while iteratting over it.
Before the keepIf helper was used that has different code for
doing this similar work.
* Cosmetics, update log and exception messages
* Update `FC` base tree updater `updateBase()`
why:
Correct `forkJunction` of canonical cursor head record. When moving
the `base`, this field would be below `base` unless updated.
* Fix `FC` chain selector `findCanonicalHead()`
why:
Given a sample ref `hash` the function searched for the unique chain
containing the block header referenced by `hash`.
Unfortunately, when searching down the ancestry lineage, the function
did not necessarily stop an the end of the sub-chain. Rather it
continued with the parent chain without noticing. So returning the
wrong result.
* When calculating new a base it must reside on cursor arc (or leg.)
why:
The finalised block argument (that will eventually be the new base)
might be moved further down the cursor arc if it is too close to the
cursor head (typically smaller than 128 blocks.)
So the finalised block selection is shifted down he cursor arc. And
it might happen that the cursor arc itself is too small and one would
end up at a parent cursor arc. This is rejected.
* Not starting a new cursor arc with a block already on another arc
why:
This leads to an inconsistent set of cursor arcs which are supposed to
be mutually disjunct.
* Tighten condition: A block that is not on the base tree must be on the DB
* One less TODO item
The EVM stack is a hot spot in EVM execution and we end up paying a nim
seq tax in several ways, adding up to ~5% of execution time:
* on initial allocation, all bytes get zeroed - this means we have to
choose between allocating a full stack or just a partial one and then
growing it
* pushing and popping introduce additional zeroing
* reallocations on growth copy + zero - expensive again!
* redundant range checking on every operation reducing inlining etc
Here a custom stack using C memory is instroduced:
* no zeroing on allocation
* full stack allocated on EVM startup -> no reallocation during
execution
* fast push/pop - no zeroing again
* 32-byte alignment - this makes it easier for the compiler to use
vector instructions
* no stack allocated for precompiles (these never use it anyway)
Of course, this change also means we have to manage memory manually -
for the EVM, this turns out to be not too bad because we already manage
database transactions the same way (they have to be freed "manually") so
we can simply latch on to this mechanism.
While we're at it, this PR also skips database lookup for known
precompiles by resolving such addresses earlier.
startedAtMs is the time passed since UNIX epoch in milliseconds.
Cannot use a monotime clock for that. As we want to keep mono
for other Moment.now() usage, imported epochTime from std/times
instead.
why:
The `base` block is ancestor to all blocks of the base tree bust stays
outside the tree.
Some fringe condition uses an opportunistic fix when the `cursor` is not in
the base tree, which is legit if `cursor != base`.
* Inline gas cost/instruction fetching
These make up 5:ish % of EVM execution time - even though they're
trivial they end up not being inlined - this little change gives a
practically free perf boost ;)
Also unify the style of creating the output to `setLen`..
* avoid a few more unnecessary seq allocations
* Ignore `FC` overlapping blocks and the ones <= `base`
why:
Due to concurrently running `importBlock()` by `newPayload` RPC
requests the `FC` module layout might differ when re-visiting for
importing blocks.
* Update logging and docu
details:
Reduce some logging noise
Clarify activating/suspending syncer in log messages
`updateOk` is obsolete and always set to true - callers should not have
to care about this detail
also take the opportunity to clean up storage root naming
* Log/trace cancellation events in scheduler
* Provide `clear()` functions for explicitly flushing data objects
* Renaming header cache functions
why:
More systematic, all functions start with prefix `dbHeader`
* Remove `danglingParent` from layout
why:
Already provided by header cache
* Remove `couplerHash` and `headHash` from layout
why:
No need to cache, `headHash` is unused and `couplerHash` used typically
once, only.
* Remove `lastLayout` from sync descriptor
why:
No need to compare changes, saving is always triggered after actively
changing the sync layout state
* Early reject unsuitable head + finalised header from CL
why:
The finalised header is only passed by its hash so the header must be
fetched somewhere, e.g. from a peer via eth/xx.
Also, finalised headers earlier than the `base` from `FC` cannot be
handled due to the `Aristo` single state database architecture.
Luckily, on a full node, the complete block history is available so
unsuitable finalised headers are stored there already which is exploited
here to avoid unnecessary network traffic.
* Code cosmetics, remove cruft, prettify logging, remove `final` metrics
detail:
The `final` layout parameter will be deprecated and later removed
* Update/re-calibrate syncer logic documentation
why:
The current implementation sucks if the `FC` module changes the
canonical branch in the middle of completing a header chain (due
to concurrent updates by the `newPayload()` logic.)
* Implement according to re-calibrated syncer docu
details:
The implementation employs the notion of named layout states (see
`SyncLayoutState` in `worker_desc.nim`) which are derived from the
state parameter triple `(C,D,H)` as described in `README.md`.
When walking AriVtx, parsing integers and nibbles actually becomes a
hotspot - these trivial changes reduces CPU usage during initial key
cache computation by ~15%.
Currently, computed hash keys are stored in a separate column family
with respect to the MPT data they're generated from - this has several
disadvantages:
* A lot of space is wasted because the lookup key (`RootedVertexID`) is
repeated in both tables - this is 30% of the `AriKey` content!
* rocksdb must maintain in-memory bloom filters and LRU caches for said
keys, doubling its "minimal efficient cache size"
* An extra disk traversal must be made to check for existence of cached
hash key
* Doubles the amount of files on disk due to each column family being
its own set of files
Here, the two CFs are joined such that both key and data is stored in
`AriVtx`. This means:
* we save ~30% disk space on repeated lookup keys
* we save ~2gb of memory overhead that can be used to cache data instead
of indices
* we can skip storing hash keys for MPT leaf nodes - these are trivial
to compute and waste a lot of space - previously they had to present in
the `AriKey` CF to avoid having to look in two tables on the happy path.
* There is a small increase in write amplification because when a hash
value is updated for a branch node, we must write both key and branch
data - previously we would write only the key
* There's a small shift in CPU usage - instead of performing lookups in
the database, hashes for leaf nodes are (re)-computed on the fly
* We can return to slightly smaller on-disk SST files since there's
fewer of them, which should reduce disk traffic a bit
Internally, there are also other advantages:
* when clearing keys, we no longer have to store a zero hash in memory -
instead, we deduce staleness of the cached key from the presence of an
updated VertexRef - this saves ~1gb of mem overhead during import
* hash key cache becomes dedicated to branch keys since leaf keys are no
longer stored in memory, reducing churn
* key computation is a lot faster thanks to the skipped second disk
traversal - a key computation for mainnet can be completed in 11 hours
instead of ~2 days (!) thanks to better cache usage and less read
amplification - with additional improvements to the on-disk format, we
can probably get rid of the initial full traversal method of seeding the
key cache on first start after import
All in all, this PR reduces the size of a mainnet database from 160gb to
110gb and the peak memory footprint during import by ~1-2gb.
Jacek Sieka
Jordan Hrycaj
andri lim
Kim De Mey
bhartnett
tersec
Advaita Saha
Kim De Mey