Skip to content

Commit eaea453

Browse files
koculukoculu
committed
Update docs.
1 parent e36676c commit eaea453

File tree

8 files changed

+27
-27
lines changed

8 files changed

+27
-27
lines changed

‎README.md

+3-3
Original file line numberDiff line numberDiff line change
@@ -45,7 +45,7 @@ Benchmark for all modes: [benchmark](https://raw.githubusercontent.com/koculu/Zo
4545
| str-str ZoneTree sync-compressed WAL | 1752 ms | 3397 ms | 5070 ms | 19153 ms |
4646
| str-str ZoneTree sync WAL | 3488 ms | 7002 ms | 10483 ms | 38727 ms |
4747
||
48-
| RocksDb sync WAL | NOT SUPPORTED |
48+
| RocksDb sync WAL (10K => 11 sec) | ~1.100.000 ms | N/A | N/A | N/A |
4949
| int-int RocksDb sync-compressed WAL | 8059 ms | 16188 ms | 23599 ms | 61947 ms |
5050
| str-str RocksDb sync-compressed WAL | 8215 ms | 16146 ms | 23760 ms | 72491 ms |
5151
||
@@ -61,12 +61,12 @@ ThresholdForMergeOperationStart = 2_000_000;
6161

6262
Additional Notes:
6363
According to our tests, ZoneTree is stable and fast even with big data.
64-
Tested up to 200M records in desktop computers till now.
64+
Tested up to 1 billion records in desktop computers till now.
6565

6666
### ZoneTree offers 4 WAL modes to let you make a flexible tradeoff.
6767

6868
* The sync mode provides maximum durability but slower write speed.
69-
In case of a crash/power cut, the sync mode ensures that the inserted data is not lost. RocksDb does not have sync WAL mode. It has a WAL mode similar to the sync-compressed mode. ( reference: [rocksdb.org](http://rocksdb.org/blog/2017/08/25/flushwal.html) )
69+
In case of a crash/power cut, the sync mode ensures that the inserted data is not lost.
7070

7171
* The sync-compressed mode provides faster write speed but less durability.
7272
Compression requires chunks to be filled before appending them into the WAL file.

‎src/ZoneTree/Core/ZoneTree.Iterators.cs

+2-2
Original file line numberDiff line numberDiff line change
@@ -143,8 +143,8 @@ public IZoneTreeIterator<TKey, TValue> CreateReadOnlySegmentsIterator(bool autoR
143143
}
144144

145145
/// <summary>
146-
/// Creates an iterator that enables scanning of the in memory segments.
147-
/// This includes readonly segments and segment zero (mutable segment).
146+
/// Creates an iterator that enables scanning of the in-memory segments.
147+
/// This includes read-only segments and mutable segment.
148148
/// </summary>
149149
/// <param name="includeDeletedRecords">if true the deleted records are included in iteration.</param>
150150
/// <returns>ZoneTree Iterator</returns>

‎src/ZoneTree/Core/ZoneTree.Merge.cs

+6-6
Original file line numberDiff line numberDiff line change
@@ -11,22 +11,22 @@ public sealed partial class ZoneTree<TKey, TValue> : IZoneTree<TKey, TValue>, IZ
1111
/// <summary>
1212
/// Moves mutable segment into readonly segment.
1313
/// This will clear the writable region of the LSM tree.
14-
/// This method is thread safe and can be called from many threads.
15-
/// The movement only occurs
16-
/// if the current segment zero is the given segment zero.
14+
/// This method is thread-safe and can be called from many threads.
15+
/// The movement only occurs if the current mutable segment
16+
/// is the mutable segment passed by argument.
1717
/// </summary>
18-
/// <param name="mutableSegment">The segment zero to move forward.</param>
18+
/// <param name="mutableSegment">The mutable segment to move forward.</param>
1919
void MoveMutableSegmentForward(IMutableSegment<TKey, TValue> mutableSegment)
2020
{
2121
lock (AtomicUpdateLock)
2222
{
2323
// move segment zero only if
24-
// the given segment zero is the current segment zero (not already moved)
24+
// the given mutable segment is the current mutable segment (not already moved)
2525
// and it is not frozen.
2626
if (mutableSegment.IsFrozen || mutableSegment != MutableSegment)
2727
return;
2828

29-
//Don't move empty segment zero.
29+
//Don't move empty mutable segment.
3030
var c = mutableSegment.Length;
3131
if (c == 0)
3232
return;

‎src/ZoneTree/IZoneTreeMaintenance.cs

+4-4
Original file line numberDiff line numberDiff line change
@@ -11,8 +11,8 @@ namespace Tenray.ZoneTree;
1111
public interface IZoneTreeMaintenance<TKey, TValue>
1212
{
1313
/// <summary>
14-
/// Gets current Segment Zero.
15-
/// Segment Zero is the writable part of the LSM tree.
14+
/// Gets current mutable segment.
15+
/// Mutable segment is the only writable part of the LSM tree.
1616
/// </summary>
1717
IMutableSegment<TKey, TValue> MutableSegment { get; }
1818

@@ -126,7 +126,7 @@ public interface IZoneTreeMaintenance<TKey, TValue>
126126
void DestroyTree();
127127

128128
/// <summary>
129-
/// Event is fired when segment zero is moved forward.
129+
/// Event is fired when mutable segment is moved forward.
130130
/// </summary>
131131
event MutableSegmentMovedForward<TKey, TValue> OnMutableSegmentMovedForward;
132132

@@ -192,7 +192,7 @@ public interface IZoneTreeMaintenance<TKey, TValue>
192192
}
193193

194194
/// <summary>
195-
/// Event is fired when segment zero is moved forward.
195+
/// Event is fired when mutable segment is moved forward.
196196
/// </summary>
197197
/// <typeparam name="TKey">The key type</typeparam>
198198
/// <typeparam name="TValue">The value type</typeparam>

‎src/ZoneTree/Options/ZoneTreeOptions.cs

+7-7
Original file line numberDiff line numberDiff line change
@@ -34,14 +34,14 @@ public class ZoneTreeOptions<TKey, TValue>
3434
/// Mutable segment maximumum key-value pair count.
3535
/// When the maximum count is reached
3636
/// MoveMutableSegmentForward is called and current mutable segment is enqueued to
37-
/// ReadOnlySegments layer.
37+
/// the ReadOnlySegments layer.
3838
/// </summary>
3939
public int MutableSegmentMaxItemCount { get; set; } = 1_000_000;
4040

4141
/// <summary>
4242
/// Disk segment maximumum key-value pair count.
4343
/// When the maximum count is reached
44-
/// The disk Segment is enqueued into to the bottom segment layer.
44+
/// The disk segment is enqueued into to the bottom segments layer.
4545
/// </summary>
4646
public int DiskSegmentMaxItemCount { get; set; } = 20_000_000;
4747

@@ -71,22 +71,22 @@ public class ZoneTreeOptions<TKey, TValue>
7171
public MarkValueDeletedDelegate<TValue> MarkValueDeleted { get; set; } = (ref TValue x) => { x = default; };
7272

7373
/// <summary>
74-
/// Write Ahead Log Options. The options is being used
75-
/// for creation of new Write Ahead Logs.
74+
/// Write Ahead Log Options. The options are used
75+
/// to create new Write Ahead Logs.
7676
/// Existing WALs is being created with their existing options.
7777
/// </summary>
7878
public WriteAheadLogOptions WriteAheadLogOptions { get; set; } = new();
7979

8080
/// <summary>
81-
/// Disk Segment options. The options is being used
82-
/// for creation of new disk segments.
81+
/// Disk Segment options. The options are used
82+
/// to create new disk segments.
8383
/// Existing disk segments is being created with
8484
/// their existing options.
8585
/// </summary>
8686
public DiskSegmentOptions DiskSegmentOptions { get; set; } = new();
8787

8888
/// <summary>
89-
/// Controls lock granularity of in memory BTree that represents
89+
/// Controls lock granularity of in-memory BTree that represents the
9090
/// mutable segment.
9191
/// </summary>
9292
public BTreeLockMode BTreeLockMode { get; set; } = BTreeLockMode.NodeLevelMonitor;

‎src/ZoneTree/docs/ZoneTree/README-NUGET.md

+3-3
Original file line numberDiff line numberDiff line change
@@ -45,7 +45,7 @@ Benchmark for all modes: [benchmark](https://raw.githubusercontent.com/koculu/Zo
4545
| str-str ZoneTree sync-compressed WAL | 1752 ms | 3397 ms | 5070 ms | 19153 ms |
4646
| str-str ZoneTree sync WAL | 3488 ms | 7002 ms | 10483 ms | 38727 ms |
4747
||
48-
| RocksDb sync WAL | NOT SUPPORTED |
48+
| RocksDb sync WAL (10K => 11 sec) | ~1.100.000 ms | N/A | N/A | N/A |
4949
| int-int RocksDb sync-compressed WAL | 8059 ms | 16188 ms | 23599 ms | 61947 ms |
5050
| str-str RocksDb sync-compressed WAL | 8215 ms | 16146 ms | 23760 ms | 72491 ms |
5151
||
@@ -61,12 +61,12 @@ ThresholdForMergeOperationStart = 2_000_000;
6161

6262
Additional Notes:
6363
According to our tests, ZoneTree is stable and fast even with big data.
64-
Tested up to 200M records in desktop computers till now.
64+
Tested up to 1 billion records in desktop computers till now.
6565

6666
### ZoneTree offers 4 WAL modes to let you make a flexible tradeoff.
6767

6868
* The sync mode provides maximum durability but slower write speed.
69-
In case of a crash/power cut, the sync mode ensures that the inserted data is not lost. RocksDb does not have sync WAL mode. It has a WAL mode similar to the sync-compressed mode. ( reference: [rocksdb.org](http://rocksdb.org/blog/2017/08/25/flushwal.html) )
69+
In case of a crash/power cut, the sync mode ensures that the inserted data is not lost.
7070

7171
* The sync-compressed mode provides faster write speed but less durability.
7272
Compression requires chunks to be filled before appending them into the WAL file.

‎src/ZoneTree/docs/ZoneTree/guide/performance.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -13,7 +13,7 @@ Benchmark for all modes: [benchmark](https://raw.githubusercontent.com/koculu/Zo
1313
| str-str ZoneTree sync-compressed WAL | 1752 ms | 3397 ms | 5070 ms | 19153 ms |
1414
| str-str ZoneTree sync WAL | 3488 ms | 7002 ms | 10483 ms | 38727 ms |
1515
||
16-
| RocksDb sync WAL | NOT SUPPORTED |
16+
| RocksDb sync WAL (10K => 11 sec) | ~1.100.000 ms | N/A | N/A | N/A |
1717
| int-int RocksDb sync-compressed WAL | 8059 ms | 16188 ms | 23599 ms | 61947 ms |
1818
| str-str RocksDb sync-compressed WAL | 8215 ms | 16146 ms | 23760 ms | 72491 ms |
1919
||

‎src/ZoneTree/docs/ZoneTree/guide/write-ahead-log.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
11
### ZoneTree offers 4 WAL modes to let you make a flexible tradeoff.
22

33
* The sync mode provides maximum durability but slower write speed.
4-
In case of a crash/power cut, the sync mode ensures that the inserted data is not lost. RocksDb does not have sync WAL mode. It has a WAL mode similar to the sync-compressed mode. ( reference: [rocksdb.org](http://rocksdb.org/blog/2017/08/25/flushwal.html) )
4+
In case of a crash/power cut, the sync mode ensures that the inserted data is not lost.
55

66
* The sync-compressed mode provides faster write speed but less durability.
77
Compression requires chunks to be filled before appending them into the WAL file.

0 commit comments

Comments
 (0)