Skip to content

Commit 615ab94

Browse files
authored
Merge pull request #6 from tidesdb/design-boards-additions
Adding some design board images to 'How does TidesDB work?'
2 parents 28520f5 + 133222b commit 615ab94

File tree

5 files changed

+8
-0
lines changed

5 files changed

+8
-0
lines changed

public/column-family-board.png

90.2 KB
Loading

public/sst-block-format-board.png

37.5 KB
Loading

public/sst-pair-merge-board.png

45.3 KB
Loading
35.5 KB
Loading

src/content/docs/getting-started/how-does-tidesdb-work.md

Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -42,6 +42,8 @@ A distinctive aspect of TidesDB is its organization around column families. Each
4242

4343
This design allows for domain-specific optimization and isolation between different types of data stored in the same database.
4444

45+
![](/column-family-board.png)
46+
4547
## 4. Core Components and Mechanisms
4648
### 4.1 Memtable
4749
The memtable is an in-memory data structure that serves as the first landing point for all write operations. Key features include:
@@ -58,6 +60,8 @@ SSTables are the immutable on-disk components of TidesDB. Their design includes:
5860
- **Min-Max Key Range**: Each SSTable stores the minimum and maximum keys it contains to optimize range queries. This block lives at block 0
5961
- **Immutability**: Once written, SSTables are never modified (only eventually merged or deleted)
6062

63+
![](/sst-block-format-board.png)
64+
6165
### 4.3 Write-Ahead Log (WAL)
6266
For durability, TidesDB implements a write-ahead logging mechanism:
6367

@@ -134,6 +138,8 @@ TidesDB implements two distinct compaction strategies:
134138
- Blocks less than manual compaction
135139
- Continues until system shutdown
136140

141+
![](/sst-pair-merge-board.png)
142+
137143
### 6.3 Compaction Mechanics
138144
During compaction:
139145

@@ -143,6 +149,8 @@ During compaction:
143149
4. Tombstones (deletion markers) and expired TTL entries are purged
144150
5. Original SSTables are deleted after successful merge
145151

152+
![](/sst-pair-merge-tombstone-board.png)
153+
146154
## 7. Performance Optimizations
147155
### 7.1 Block Indices
148156
TidesDB employs block indices to optimize read performance:

0 commit comments

Comments
 (0)