Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merge tree service in master #1607

Merged
merged 46 commits into from
Jul 21, 2022
Merged

Merge tree service in master #1607

merged 46 commits into from
Jul 21, 2022

Conversation

fyrchik
Copy link
Contributor

@fyrchik fyrchik commented Jul 18, 2022

Only the last 2 commits are fresh.
Tree service must now be explicitly enabled via tree.enabled config field.

@codecov
Copy link

codecov bot commented Jul 18, 2022

Codecov Report

Merging #1607 (9f1c597) into master (198beae) will decrease coverage by 2.79%.
The diff coverage is 20.28%.

❗ Current head 9f1c597 differs from pull request most recent head d537e9e. Consider uploading reports for the commit d537e9e to get more accurate results

@@            Coverage Diff             @@
##           master    #1607      +/-   ##
==========================================
- Coverage   35.92%   33.13%   -2.80%     
==========================================
  Files         305      329      +24     
  Lines       18464    22468    +4004     
==========================================
+ Hits         6633     7444     +811     
- Misses      11273    14421    +3148     
- Partials      558      603      +45     
Impacted Files Coverage Δ
pkg/local_object_storage/engine/tree.go 0.00% <0.00%> (ø)
pkg/local_object_storage/pilorama/info.go 0.00% <0.00%> (ø)
pkg/local_object_storage/shard/info.go 0.00% <ø> (ø)
pkg/local_object_storage/shard/tree.go 0.00% <0.00%> (ø)
pkg/services/control/convert.go 0.00% <0.00%> (ø)
pkg/services/control/rpc.go 0.00% <0.00%> (ø)
pkg/services/control/service.go 21.97% <0.00%> (-1.56%) ⬇️
pkg/services/control/service_grpc.pb.go 0.00% <0.00%> (ø)
pkg/services/tree/cache.go 0.00% <0.00%> (ø)
pkg/services/tree/container.go 0.00% <0.00%> (ø)
... and 26 more

cmd/neofs-node/tree.go Outdated Show resolved Hide resolved
carpawell
carpawell previously approved these changes Jul 18, 2022
@fyrchik
Copy link
Contributor Author

fyrchik commented Jul 18, 2022

Updated only 2 fixes for the linter.

carpawell
carpawell previously approved these changes Jul 19, 2022
@fyrchik
Copy link
Contributor Author

fyrchik commented Jul 19, 2022

Fixed conflicts with master.

carpawell
carpawell previously approved these changes Jul 19, 2022
fyrchik and others added 23 commits July 21, 2022 13:43
In this commit we implement algorithm for CRDT trees from
https://martin.klepmann.com/papers/move-op.pdf

Each tree is identified by the ID of a container it belongs to
and the tree name itself. Essentially, it is a sequence of operations
which should be applied in chronological order to get a usual tree
representation.

There are 2 backends for now: bbolt database and in-memory.
In-memory backend is here for debugging and will eventually act
as a memory-cache for the on-disk database.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Object Tree Service allows changing trees assotiated with
the container in runtime.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Denis Kirillov <denis@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Consider a node `{FileName: "dir", Attribute: "xxx"}`. In case we add
a new node by path `["dir", "file.txt"]`, create a new intermediate node
with a single attribute.

`GetByPath` now also considers only nodes with a single attribute while building a path.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Currently to find a node by path we iterate over all the children on
each level. This is far from optimal and scales badly with the number of
nodes on a single level. Thus we introduce "indexed attributes" for
which an additional information is stored and which can be use in
`*ByPath` operations. Currently this set only includes `FileName`
attribute but this may change in future.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Helps a lot in case of concurrent request flow.

```
name                      old time/op    new time/op    delta
ApplySequential/bbolt-8     78.0µs ± 9%    59.8µs ± 4%  -23.39%  (p=0.000 n=10+9)
ApplyReorderLast/bbolt-8     143µs ± 5%     113µs ±15%  -21.06%  (p=0.000 n=10+10)

name                      old alloc/op   new alloc/op   delta
ApplySequential/bbolt-8     56.9kB ± 8%    28.9kB ± 3%  -49.22%  (p=0.000 n=10+10)
ApplyReorderLast/bbolt-8    87.3kB ± 3%    40.9kB ±10%  -53.16%  (p=0.000 n=10+10)

name                      old allocs/op  new allocs/op  delta
ApplySequential/bbolt-8        224 ±11%       262 ± 5%  +16.93%  (p=0.000 n=9+10)
ApplyReorderLast/bbolt-8       518 ± 4%       674 ±11%  +30.09%  (p=0.000 n=10+10)
```

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Also fix a bug with replicator using the multiaddress instead of
<host>:<port> format expected by gRPC library.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
…ntainer

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
```
name                      old time/op    new time/op    delta
ApplySequential/bbolt-8     55.5µs ± 4%    55.5µs ± 3%     ~     (p=1.000 n=10+7)
ApplyReorderLast/bbolt-8     108µs ± 6%     112µs ± 8%     ~     (p=0.077 n=9+9)

name                      old alloc/op   new alloc/op   delta
ApplySequential/bbolt-8     28.8kB ± 3%    27.7kB ± 6%   -3.79%  (p=0.005 n=10+10)
ApplyReorderLast/bbolt-8    41.4kB ± 5%    38.9kB ± 5%   -6.19%  (p=0.001 n=10+9)

name                      old allocs/op  new allocs/op  delta
ApplySequential/bbolt-8        262 ± 2%       235 ±10%  -10.41%  (p=0.000 n=10+10)
ApplyReorderLast/bbolt-8       684 ± 6%       616 ± 7%  -10.04%  (p=0.000 n=10+9)
```

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Before this commit the replication channel was quickly filled under
heavy load. This lead to the continuously increasing latency for all
write operations. Now it looks better.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
… TLS

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
In case node is down or failing for some reason, we can expect `Dial` to
fail. In case we actively try to replicate and `Dial` always takes 2
seconds, replication-related channels quickly become full. That affects
latency of all other write operations.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Current implementation prevents invalid operations to become valid at
some later point (consider adding a child to the non-existent parent and
then adding the parent). This seems to diverge from the paper algorithm
and complicates implementation. Make it simpler.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
The tricky part here is the engine itself: we stop iteration on
`ErrReadOnly` because it is better to synchronize the shard later than
to have partial trees stored in 2 shards.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Do not return backend type from the service for now, because memory
backend is expected to vanish.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
1. Modifying operations are not expected to fail, unless the shard is
   read-only.
2. `Get*` operations should increase error counter too, unless the
   error is `ErrTreeNotFound`.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Do not check that a node indeed belongs to the container, because the
synchronization will fail in this case anyway.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
@fyrchik fyrchik merged commit 2455b72 into master Jul 21, 2022
aprasolova pushed a commit to aprasolova/neofs-node that referenced this pull request Oct 19, 2022
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
aprasolova pushed a commit to aprasolova/neofs-node that referenced this pull request Oct 19, 2022
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
aprasolova pushed a commit to aprasolova/neofs-node that referenced this pull request Oct 19, 2022
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
aprasolova pushed a commit to aprasolova/neofs-node that referenced this pull request Oct 19, 2022
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants