https://github.com/openinx/minibase

An embedded KV storage engine for learning HBase
https://github.com/openinx/minibase

Last synced: 15 days ago
JSON representation

An embedded KV storage engine for learning HBase

• Host: GitHub
• URL: https://github.com/openinx/minibase
• Owner: openinx
• Created: 2019-01-20T09:49:30.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2020-10-13T11:42:27.000Z (about 4 years ago)
• Last Synced: 2024-06-08T12:35:44.417Z (5 months ago)
• Language: Java
• Size: 39.1 KB
• Stars: 237
• Watchers: 7
• Forks: 144
• Open Issues: 6
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
MiniBase is an embedded KV storage engine, it's quit simple, not for production env, just for better understand HBase or

other LSM-related index algorithm.

In MiniBase, we use those basic algorithm and data structure:

* BloomFilter: it can helps a lot when filtering much useless IO.

* ConcurrentSkipListMap: Yeah, it's quite suitable when designing memstore. It can maintian an sorted key value set in

high concurrency scenarios.

* LSM Index Algorithm: the memstore part and disk store part.

### How to use ?

```java

Config conf = new Config().setDataDir(dataDir).setMaxMemstoreSize(1).setFlushMaxRetries(1)

    .setMaxDiskFiles(10);

MiniBase db = MiniBaseImpl.create(conf).open();

// Put

db.put(Bytes.toBytes(1), Bytes.toBytes(1));

// Scan

Iter kv = db.scan();

while (kv.hasNext()) {

    KeyValue kv = kv.next();

    //...

}

```

### How to build and test?

```shell

git clone [email protected]:openinx/minibase.git

mvn clean package

```