Commit graph

111 commits

Author SHA1 Message Date
Roman Khimov
9576e10d04 storage: use map size hints to optimize subsequent persist()
We're likely to have something comparable to the current changeset in the
subsequent one. If it's bigger, no big deal, it'll be reallocated, if it's
smaller, no big deal, the next one will be preallocated smaller.
2021-12-01 21:36:25 +03:00
Roman Khimov
33e37e60e5
Merge pull request #2264 from nspcc-dev/fix-win-tests
*: Windows compatibility fixes
2021-11-29 11:25:35 +03:00
AnnaShaleva
aefb6f9fee *: fix tests failing due to path.Join usage
Solution:
Use `file/filepath` package to construct expected path. This package is OS-aware, see https://github.com/golang/go/issues/30616.
2021-11-29 11:11:09 +03:00
AnnaShaleva
06beb4d534 core: fix TestMemCachedPersist test
Problem:
```
--- FAIL: TestMemCachedPersist (0.07s)
    --- FAIL: TestMemCachedPersist/BoltDBStore (0.07s)
        testing.go:894: TempDir RemoveAll cleanup: remove C:\Users\Anna\AppData\Local\Temp\TestMemCachedPersist_BoltDBStore294966711\001\test_bolt_db: The process cannot access the file because it is being used by another process.
```

Solution:
Release the resources occupied by the DB.
2021-11-26 18:26:27 +03:00
Evgeniy Stratonikov
fac595bbdf core: remove old storage items asynchronously
Signed-off-by: Evgeniy Stratonikov <evgeniy@nspcc.ru>
2021-11-24 16:25:11 +03:00
Roman Khimov
5fbe838fd4 core: add and use synchronous persist to avoid OOM
b9be892bf9 has made Persist asynchronous which
is very effective in allowing the system to continue processing
blocks/transactions while flushing things to disk. It at the same time is very
dangerous in that if the disk is slow and it takes much time to flush KV set
(more than persisting interval), there might be even bigger new KV set in
MemCachedStore by the time it finishes. Even if the system immediately starts
to flush this new data set it (being bigger) can take more time than the
previous one. And while doing so a new data set will appear in memory,
potentially again bigger than this.

So we can easily end up with the system going out of control, consuming more
and more memory and taking more and more time to persist a single set of
data. To avoid this we need to detect such condition and just wait for Persist
to really finish its job and release the resources.
2021-11-22 10:41:40 +03:00
Roman Khimov
125e4231b0 core: store NEP-11 transfers, add accessor functions 2021-11-18 00:09:10 +03:00
Roman Khimov
1144a03486 storage: drop RedisDB, close #2130 2021-10-27 17:32:25 +03:00
Roman Khimov
fb4b87bb96 storage: drop BadgerDB support, close #2130 2021-10-27 17:31:55 +03:00
Anna Shaleva
3450371910 core: split (*MemCachedStore) Seek and SeekAsync methods
Use SeekAsync for System.Storage.Find and Seek for the rest of cases.
2021-10-21 10:05:12 +03:00
Anna Shaleva
dcda7bec63 core: squash PS-seeking and merging routines in (*MemCachedStore).Seek
We don't need a separate routine to merge seek results.
2021-10-21 10:05:12 +03:00
Anna Shaleva
c88720bf45 core: remove memstore routine from (*MemCachedStore).SeekAsync
It adds unnecessary overhead to computations.
2021-10-21 10:05:12 +03:00
Anna Shaleva
dfe2c667e1 core: do not hold the lock while seeking over persistent store 2021-10-21 10:05:12 +03:00
Anna Shaleva
0a4f45c9b0 core: add ability to free storage.Iterator resources 2021-10-21 10:05:12 +03:00
Anna Shaleva
89ee2e7720 core: refactor storage.Find and storage.Iterator to work with channel
Add SeekAsync methods in order to fetch matching storage items
on demand. Refactor storage.Find and storage.Iterator wrt these changes.
2021-10-21 10:05:12 +03:00
Anna Shaleva
7ba88e98e2 core: optimize (*MemCachedStore).Seek operation
Real persistent storage guarantees that result of Seek is sorted
by keys. The idea of optimisation is to merge two sorted seek
results into one (memStore+persistentStore), so that
(*MemCachedStore).Seek will return sorted list. The only thing
that remains is to sort items got from (*MemoryStore).Seek.
2021-10-21 10:05:12 +03:00
Anna Shaleva
191cc45032 core: sort items in MemoryStore.Seek
MemoryStore is used in a MemCachedStore as a persistent layer in tests.
Further commits suppose that persistent storage returns sorted values
from Seek, so sort the result of MemoryStore.Seek.

Benchmark results for 10000 matching items in MemoryStore compared to
master:
name          old time/op    new time/op    delta
MemorySeek-8     712µs ± 0%    3850µs ± 0%   +440.52%  (p=0.000 n=8+8)

name          old alloc/op   new alloc/op   delta
MemorySeek-8     160kB ± 0%    2724kB ± 0%  +1602.61%  (p=0.000 n=10+8)

name          old allocs/op  new allocs/op  delta
MemorySeek-8     10.0k ± 0%     10.0k ± 0%     +0.24%  (p=0.000 n=10+10)

For details on implementation efficiency see the
https://github.com/nspcc-dev/neo-go/pull/2193#discussion_r722993358.
2021-10-21 10:05:12 +03:00
Anna Shaleva
d8210c0137 core: add benchmarks for iterator.Next, MemCached.Seek, Mem.Seek 2021-10-21 10:05:12 +03:00
Anna Shaleva
8d8071f97e core: distinguish storage.KeyValue and storage.KeyValueExists
We need Exists field for storage batch related code; other cases may go
without Exists, so add new KeyValue structure and refactor related code.
2021-10-21 10:05:12 +03:00
Anna Shaleva
5cd78c31af core: allow to recover after state jump interruption
We need several stages to manage state jump process in order not to mess
up old and new contract storage items and to be sure about genesis state data
are properly removed from the storage. Other operations do not require
separate stage and can be performed each time `jumpToStateInternal` is
called.
2021-09-07 19:43:27 +03:00
Anna Shaleva
6381173293 core: store statesync-related storage items under temp prefix
State jump should be an atomic operation, we can't modify contract
storage items state on-the-fly. Thus, store fresh items under temp
prefix and replase the outdated ones after state sync is completed.
Related
https://github.com/nspcc-dev/neo-go/pull/2019#discussion_r693350460.
2021-09-07 19:43:27 +03:00
Roman Khimov
6d074a96e9 *: make tests use TempDir(), fix #1319
Simplify things, drop TempFile at the same time (refs. #1764)
2021-08-26 17:29:40 +03:00
Roman Khimov
adc660c3e0
Merge pull request #2123 from nspcc-dev/store-better
Store better
2021-08-13 12:50:24 +03:00
Roman Khimov
ae071d4542 storage: introduce PutChangeSet and use it for Persist
We're using batches in wrong way during persist, we already have all changes
accumulated in two maps and then we move them to batch and then this is
applied. For some DBs like BoltDB this batch is just another MemoryStore, so
we essentially just shuffle the changeset from one map to another, for others
like LevelDB batch is just a serialized set of KV pairs, it doesn't help much
on subsequent PutBatch, we just duplicate the changeset again.

So introduce PutChangeSet that allows to take two maps with sets and deletes
directly. It also allows to simplify MemCachedStore logic.

neo-bench for single node with 10 workers, LevelDB:

  Reference:

  RPS    30189.132 30556.448 30390.482 ≈ 30379    ±  0.61%
  TPS    29427.344 29418.687 29434.273 ≈ 29427    ±  0.03%
  CPU %     33.304    27.179    33.860 ≈    31.45 ± 11.79%
  Mem MB   800.677   798.389   715.042 ≈   771    ±  6.33%

  Patched:

  RPS    30264.326 30386.364 30166.231 ≈ 30272    ± 0.36% ⇅
  TPS    29444.673 29407.440 29452.478 ≈ 29435    ± 0.08% ⇅
  CPU %     34.012    32.597    33.467 ≈   33.36  ± 2.14% ⇅
  Mem MB   549.126   523.656   517.684 ≈  530     ± 3.15% ↓ 31.26%

BoltDB:

  Reference:

  RPS    31937.647 31551.684 31850.408 ≈ 31780    ±  0.64%
  TPS    31292.049 30368.368 31307.724 ≈ 30989    ±  1.74%
  CPU %     33.792    22.339    35.887 ≈    30.67 ± 23.78%
  Mem MB  1271.687  1254.472  1215.639 ≈  1247    ±  2.30%

  Patched:

  RPS    31746.818 30859.485 31689.761 ≈ 31432    ± 1.58% ⇅
  TPS    31271.499 30340.726 30342.568 ≈ 30652    ± 1.75% ⇅
  CPU %     34.611    34.414    31.553 ≈    33.53 ± 5.11% ⇅
  Mem MB  1262.960  1231.389  1335.569 ≈  1277    ± 4.18% ⇅
2021-08-12 17:42:16 +03:00
Roman Khimov
18682f2409 storage: don't use locks for memory batches
They're inherently single-threaded, so locking makes no sense for them.
2021-08-11 18:55:07 +03:00
Anna Shaleva
0e3b9c48a2 core: add API to store StateSyncPoint and StateSyncCurrentBlockHeight
We need it in order not to mess up the blockchain which has its own
CurrentBlockHeight.
2021-08-10 14:06:28 +03:00
Roman Khimov
b9be892bf9 storage: allow accessing MemCachedStore during Persist
Persist by its definition doesn't change MemCachedStore visible state, all KV
pairs that were acessible via it before Persist remain accessible after
Persist. The only thing it does is flushing of the current set of KV pairs
from memory to peristent store. To do that it needs read-only access to the
current KV pair set, but technically it then replaces maps, so we have to use
full write lock which makes MemCachedStore inaccessible for the duration of
Persist. And Persist can take a lot of time, it's about disk access for
regular DBs.

What we do here is we create new in-memory maps for MemCachedStore before
flushing old ones to the persistent store. Then a fake persistent store is
created which actually is a MemCachedStore with old maps, so it has exactly
the same visible state. This Store is never accessed for writes, so we can
read it without taking any internal locks and at the same time we no longer
need write locks for original MemCachedStore, we're not using it. All of this
makes it possible to use MemCachedStore as normally reads are handled going
down to whatever level is needed and writes are handled by new maps. So while
Persist for (*Blockchain).dao does its most time-consuming work we can process
other blocks (reading data for transactions and persisting storeBlock caches
to (*Blockchain).dao).

The change was tested for performance with neo-bench (single node, 10 workers,
LevelDB) on two machines and block dump processing (RC4 testnet up to 62800
with VerifyBlocks set to false) on i7-8565U.

Reference results (bbe4e9cd7b):

Ryzen 9 5950X:
RPS     23616.969 22817.086 23222.378  ≈ 23218   ± 1.72%
TPS     23047.316 22608.578 22735.540  ≈ 22797   ± 0.99%
CPU %      23.434    25.553    23.848  ≈    24.3 ± 4.63%
Mem MB    600.636   503.060   582.043  ≈   562   ± 9.22%

Core i7-8565U:
RPS     6594.007 6499.501 6572.902  ≈ 6555   ± 0.76%
TPS     6561.680 6444.545 6510.120  ≈ 6505   ± 0.90%
CPU %     58.452   60.568   62.474    ≈ 60.5 ± 3.33%
Mem MB   234.893  285.067  269.081   ≈ 263   ± 9.75%

DB restore:
real    0m22.237s 0m23.471s 0m23.409s  ≈ 23.04 ± 3.02%
user    0m35.435s 0m38.943s 0m39.247s  ≈ 37.88 ± 5.59%
sys      0m3.085s  0m3.360s  0m3.144s  ≈  3.20 ± 4.53%

After the change:

Ryzen 9 5950X:
RPS     27747.349 27407.726 27520.210  ≈ 27558   ± 0.63%  ↑ 18.69%
TPS     26992.010 26993.468 27010.966  ≈ 26999   ± 0.04%  ↑ 18.43%
CPU %      28.928    28.096    29.105  ≈    28.7 ± 1.88%  ↑ 18.1%
Mem MB    760.385   726.320   756.118  ≈   748   ± 2.48%  ↑ 33.10%

Core i7-8565U:
RPS     7783.229 7628.409 7542.340  ≈ 7651   ± 1.60%  ↑ 16.72%
TPS     7708.436 7607.397 7489.459  ≈ 7602   ± 1.44%  ↑ 16.85%
CPU %     74.899   71.020   72.697  ≈   72.9 ± 2.67%  ↑ 20.50%
Mem MB   438.047  436.967  416.350  ≈  430   ± 2.84%  ↑ 63.50%

DB restore:
real    0m20.838s 0m21.895s 0m21.794s  ≈ 21.51 ± 2.71%  ↓ 6.64%
user    0m39.091s 0m40.565s 0m41.493s  ≈ 40.38 ± 3.00%  ↑ 6.60%
sys      0m3.184s  0m2.923s  0m3.062s  ≈  3.06 ± 4.27%  ↓ 4.38%

It obviously uses more memory now and utilizes CPU more aggressively, but at
the same time it allows to improve all relevant metrics and finally reach a
situation where we process 50K transactions in less than second on Ryzen 9
5950X (going higher than 25K TPS). The other observation is much more stable
block time, on Ryzen 9 it's as close to 1 second as it could be.
2021-08-02 16:33:00 +03:00
Anna Shaleva
e46d76d7aa core: rename state.NEP17Balances to state.NEP17TransferInfo
Balances are to be removed from state.NEP17TransferInfo, so the remnant
fields are NextTransferBatch, NewBatch and a map of LastUpdatedBlocks.
These fields are more staff-related.

Also rename dao.[Get, Put, put]NEP17Balances and STNEP17Balances
preffix.

Also rename NEP17TransferInfo.Trackers to LastUpdatedBlockTrackers
because NEP17TransferInfo.Balances are to be removed.
2021-07-28 13:22:53 +03:00
Roman Khimov
002ad9dfee go.mod: update miniredis to 2.15.1
It's only used for testing purposes and this version doesn't change anything
for us, but still better be current.
2021-07-21 23:28:26 +03:00
Roman Khimov
3b19b34122 storage: fix memcached test with boltdb store
Everything was wrong here, wrong file used, wrong cleanup procedure, the net
result is this (and some failing tests from time to time):

  $ ls -l /tmp/test_bolt_db* | wc -l
  30939
2021-07-20 12:35:24 +03:00
Roman Khimov
19717dd9a8 slice: introduce common Copy helper
It's a bit more convenient to use.
2021-07-19 22:57:55 +03:00
Evgeniy Stratonikov
9f377cde12 storage: convert key once in MemoryStore.seek
There is no need in additional allocations.

Signed-off-by: Evgeniy Stratonikov <evgeniy@nspcc.ru>
2021-07-12 14:14:32 +03:00
Roman Khimov
9d2712573f *: enable godot linter and fix all its warnings
It's important for NeoGo to have clean documentation. No functional changes.
2021-05-12 23:17:03 +03:00
Anna Shaleva
f0c222b385 core: move StorageItem-related constants to storage pkg
Need this to avoid import cycle problem.
2021-04-01 10:17:08 +03:00
Anna Shaleva
2c81fc8b8e *: upgrade tests to use T.Cleanup() 2021-03-01 17:08:00 +03:00
Evgenii Stratonikov
30423f3306 mpt: update MPT after the block processing 2020-12-25 14:40:23 +03:00
Roman Khimov
ab12eee346 native: move contract deployment to management contract
See neo-project/neo#2119.
2020-12-14 15:23:46 +03:00
Evgenii Stratonikov
31eca342eb *: replace all NEP5 occurences to NEP17 2020-11-24 13:08:24 +03:00
Roman Khimov
0f39da19d0 storage: fix Get for BoltDB, fix #1482 2020-10-13 19:05:11 +03:00
Anna Shaleva
097ef2a731 core: copy storage item key in simple.GetStorageItems
Related to #1468, ported from #1475.

We should copy the key to avoid bytes substitution. Otherwise there's a
chance that at the end of dao.Store.Seek(...) execution some keys won't
be the same as the original keys found inside saveToMap function because
storage.Seek can guarantee that provided key and value are only valid
until the next `f` call.
2020-10-10 12:37:46 +03:00
Roman Khimov
bc31ab3d2c storage: add bloom filter to leveldb
We're constantly checking for transactions there and most of the time this
check is not successful (meaning that the transaction in question is
new). Bloom filter easily reduces the need to search over the DB in 99% of
these cases and gives some 13% increase in single-node TPS.
2020-09-09 20:46:31 +03:00
Anna Shaleva
fdf73597af core: return error for unknown storage configuration 2020-08-25 16:11:44 +03:00
Roman Khimov
0e2784cd2c always wrap errors when creating new ones with fmt.Errorf()
It doesn't really change anything in most of the cases, but it's a useful
habit anyway.

Fix #350.
2020-08-07 12:21:52 +03:00
fyrchik
e4fc655115
Merge pull request #1239 from nspcc-dev/store_nep5_with_id
core: store contract IDs instead of hashes for NEP5Balances and Transfers
2020-07-30 12:56:19 +03:00
Evgenii Stratonikov
dc6741bce7 mpt: implement MPT trie
MPT is a trie with a branching factor = 16, i.e. it consists of sequences in
16-element alphabet.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2020-07-30 12:42:15 +03:00
Anna Shaleva
dbd460d883 core: retrieve contract hash by ID
We'll need this ability further to retrieve contracts hashes for Nep5Balances.
2020-07-29 15:14:24 +03:00
Evgenii Stratonikov
a03af55732 core: provide monotonically increasing ID for contracts
ID is a contract identifier which doesn't change during migration.
It is also used as a part of storage context instead of contract hash.
2020-06-11 10:45:25 +03:00
Roman Khimov
a986e2a064 *: drop support for old on-chain assets
You no longer can transfer them, so creating/renewing/storing doesn't make
much sense.
2020-06-05 19:21:37 +03:00
Roman Khimov
bf6435eeaa storage: drop STValidator prefix, it's not used 2020-06-05 19:20:16 +03:00
Roman Khimov
709146f295 transaction: drop Inputs and Outputs, forget UTXO 2020-06-05 19:20:16 +03:00