NutsDB

English | 简体中文

NutsDB is a simple, fast, embeddable and persistent key/value store written in pure Go.

It supports fully serializable transactions and many data structures such as list、set、sorted set. All operations happen inside a Tx. Tx represents a transaction, which can be read-only or read-write. Read-only transactions can read values for a given bucket and a given key or iterate over a set of key-value pairs. Read-write transactions can read, update and delete keys from the DB.

Motivation

I wanted a simple, fast, embeddable and persistent key/value store written in pure Go. And if it supports more data structures such as list, set, sorted set，it will be better.

There are some options around the embeddable key/value store in Go:

BoltDB is based on B+ tree, has a good random read performance and awesome sequential scan performance, and it supports ACID transactions with serializable isolation, but it is terrible at random write performance and not supports more data structures such as list, etc.

GoLevelDB is based on a log-structured merge-tree (LSM tree), but it does not support more data structures.

Badger is based on LSM tree with value log. It designed for SSDs. It also supports transactions. But its write performance is not as good as I thought. And it also does not support more data structures.

Moreover, I was curious about how to implement a key/value database. The database can be said to be the core of the system, to understand the database kernel or their own implementation, better use of the same kind of database or the next time according to the business custom database is very helpful.

So I tried to build a key/value store by myself, I wanted to find a simple store engine model as a reference. Finally, I found the Bitcask model. It is simple and easy to implement. However, it has its limitation, like range or prefix queries, are not efficient. For example, you cannot easily scan over all keys between user000000 and user999999, you had to look up each key individually in the hashmap.

In order to break the limitation, I tried to optimize them. Finally, I did it and named NutsDB. NutsDB offers a high read/write performance and supports transactions. And it still has a lot of room for optimization. Welcome contributions to NutsDB.

Table of Contents

• Getting Started
  • Installing
  • Opening a database
  • Options
    • Default Options
  • Transactions
    • Read-write transactions
    • Read-only transactions
    • Managing transactions manually
  • Using buckets
  • Using key/value pairs
  • Using TTL(Time To Live)
  • Iterating over keys
    • Prefix scans
    • Prefix search scans
    • Range scans
    • Get all
  • Merge Operation
  • Database backup
• Using Other data structures
  • List
    • RPush
    • LPush
    • LPop
    • LPeek
    • RPop
    • RPeek
    • LRange
    • LRem
    • LSet
    • Ltrim
    • LSize
  • Set
    • SAdd
    • SAreMembers
    • SCard
    • SDiffByOneBucket
    • SDiffByTwoBuckets
    • SHasKey
    • SIsMember
    • SMembers
    • SMoveByOneBucket
    • SMoveByTwoBuckets
    • SPop
    • SRem
    • SUnionByOneBucket
    • SUnionByTwoBucket
  • Sorted Set
    • ZAdd
    • ZCard
    • ZCount
    • ZGetByKey
    • ZMembers
    • ZPeekMax
    • ZPeekMin
    • ZPopMax
    • ZPopMin
    • ZRangeByRank
    • ZRangeByScore
    • ZRank
    • ZRevRank
    • ZRem
    • ZRemRangeByRank
    • ZScore
• Comparison with other databases
  • BoltDB
  • LevelDB, RocksDB
  • Badger
• Benchmarks
• Caveats & Limitations
• Contact
• Contributing
• Acknowledgements
• License

Getting Started

Installing

To start using NutsDB, first needs Go installed (version 1.11+ is required). and run go get:

go get -u github.com/xujiajun/nutsdb

Opening a database

To open your database, use the nutsdb.Open() function,with the appropriate options.The Dir , EntryIdxMode and SegmentSize options are must be specified by the client. About options see here for detail.

package main

import (
	"log"

	"github.com/xujiajun/nutsdb"
)

func main() {
	// Open the database located in the /tmp/nutsdb directory.
	// It will be created if it doesn't exist.
	opt := nutsdb.DefaultOptions
	opt.Dir = "/tmp/nutsdb"
	db, err := nutsdb.Open(opt)
	if err != nil {
		log.Fatal(err)
	}
	defer db.Close()

	...
}

Options

• Dir string

Dir represents Open the database located in which dir.

• EntryIdxMode EntryIdxMode

EntryIdxMode represents using which mode to index the entries. EntryIdxMode includes three options: HintKeyValAndRAMIdxMode,HintKeyAndRAMIdxMode and HintBPTSparseIdxMode. HintKeyValAndRAMIdxMode represents ram index (key and value) mode, HintKeyAndRAMIdxMode represents ram index (only key) mode and HintBPTSparseIdxMode represents b+ tree sparse index mode.

• RWMode RWMode

RWMode represents the read and write mode. RWMode includes two options: FileIO and MMap. FileIO represents the read and write mode using standard I/O. And MMap represents the read and write mode using mmap.

• SegmentSize int64

NutsDB will truncate data file if the active file is larger than SegmentSize. Current verison default SegmentSize is 8MB,but you can custom it. Once set, it cannot be changed. see caveats--limitations for detail.

• NodeNum int64

NodeNum represents the node number.Default NodeNum is 1. NodeNum range [1,1023] .

• SyncEnable bool

SyncEnable represents if call Sync() function. if SyncEnable is false, high write performance but potential data loss likely. if SyncEnable is true, slower but persistent.

• StartFileLoadingMode RWMode

StartFileLoadingMode represents when open a database which RWMode to load files.

Default Options

Recommend to use the DefaultOptions . Unless you know what you're doing.

var DefaultOptions = Options{
	EntryIdxMode:         HintKeyValAndRAMIdxMode,
	SegmentSize:          defaultSegmentSize,
	NodeNum:              1,
	RWMode:               FileIO,
	SyncEnable:           true,
	StartFileLoadingMode: MMap,
}

Transactions

NutsDB allows only one read-write transaction at a time but allows as many read-only transactions as you want at a time. Each transaction has a consistent view of the data as it existed when the transaction started.

When a transaction fails, it will roll back, and revert all changes that occurred to the database during that transaction. if set the option SyncEnable true When a read/write transaction succeeds all changes are persisted to disk.

Creating transaction from the DB is thread safe.

Read-write transactions

err := db.Update(
	func(tx *nutsdb.Tx) error {
	...
	return nil
})

Read-only transactions

err := db.View(
	func(tx *nutsdb.Tx) error {
	...
	return nil
})

Managing transactions manually

The DB.View() and DB.Update() functions are wrappers around the DB.Begin() function. These helper functions will start the transaction, execute a function, and then safely close your transaction if an error is returned. This is the recommended way to use NutsDB transactions.

However, sometimes you may want to manually start and end your transactions. You can use the DB.Begin() function directly but please be sure to close the transaction.

 // Start a write transaction.
tx, err := db.Begin(true)
if err != nil {
    return err
}

bucket := "bucket1"
key := []byte("foo")
val := []byte("bar")

// Use the transaction.
if err = tx.Put(bucket, key, val, Persistent); err != nil {
	// Rollback the transaction.
	tx.Rollback()
} else {
	// Commit the transaction and check for error.
	if err = tx.Commit(); err != nil {
		tx.Rollback()
		return err
	}
}

Using buckets

Buckets are collections of key/value pairs within the database. All keys in a bucket must be unique. Bucket can be interpreted as a table or namespace. So you can store the same key in different bucket.

key := []byte("key001")
val := []byte("val001")

bucket001 := "bucket001"
if err := db.Update(
	func(tx *nutsdb.Tx) error {
		if err := tx.Put(bucket001, key, val, 0); err != nil {
			return err
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

bucket002 := "bucket002"
if err := db.Update(
	func(tx *nutsdb.Tx) error {
		if err := tx.Put(bucket002, key, val, 0); err != nil {
			return err
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

Also, this bucket is related to the data structure you use. Different data index structures that use the same bucket are also different. For example, you define a bucket named bucket_foo, so you need to use the list data structure, use tx.RPush to add data, you must query or retrieve from this bucket_foo data structure, use tx.RPop, tx.LRange, etc. You cannot use tx.Get (same index type as tx.GetAll, tx.Put, tx.Delete, tx.RangeScan, etc.) to read the data in this bucket_foo, because the index structure is different. Other data structures such as Set, Sorted Set are the same.

Using key/value pairs

To save a key/value pair to a bucket, use the tx.Put method:

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	key := []byte("name1")
	val := []byte("val1")
	bucket := "bucket1"
	if err := tx.Put(bucket, key, val, 0); err != nil {
		return err
	}
	return nil
}); err != nil {
	log.Fatal(err)
}

This will set the value of the "name1" key to "val1" in the bucket1 bucket.

To update the the value of the "name1" key,we can still use the tx.Put function:

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	key := []byte("name1")
	val := []byte("val1-modify") // Update the value
	bucket := "bucket1"
	if err := tx.Put(bucket, key, val, 0); err != nil {
		return err
	}
	return nil
}); err != nil {
	log.Fatal(err)
}

To retrieve this value, we can use the tx.Get function:

if err := db.View(
func(tx *nutsdb.Tx) error {
	key := []byte("name1")
	bucket := "bucket1"
	if e, err := tx.Get(bucket, key); err != nil {
		return err
	} else {
		fmt.Println(string(e.Value)) // "val1-modify"
	}
	return nil
}); err != nil {
	log.Println(err)
}

Use the tx.Delete() function to delete a key from the bucket.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	key := []byte("name1")
	bucket := "bucket1"
	if err := tx.Delete(bucket, key); err != nil {
		return err
	}
	return nil
}); err != nil {
	log.Fatal(err)
}

Using TTL(Time To Live)

NusDB supports TTL(Time to Live) for keys, you can use tx.Put function with a ttl parameter.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	key := []byte("name1")
	val := []byte("val1")
	bucket := "bucket1"
	
	// If set ttl = 0 or Persistent, this key will nerver expired.
	// Set ttl = 60 , after 60 seconds, this key will expired.
	if err := tx.Put(bucket, key, val, 60); err != nil {
		return err
	}
	return nil
}); err != nil {
	log.Fatal(err)
}

Iterating over keys

NutsDB stores its keys in byte-sorted order within a bucket. This makes sequential iteration over these keys extremely fast.

Prefix scans

To iterate over a key prefix, we can use PrefixScan function, and the parameters offsetNum and limitNum constrain the number of entries returned :

if err := db.View(
	func(tx *nutsdb.Tx) error {
		prefix := []byte("user_")
		bucket := "user_list"
		// Constrain 100 entries returned 
		if entries, _, err := tx.PrefixScan(bucket, prefix, 25, 100); err != nil {
			return err
		} else {
			for _, entry := range entries {
				fmt.Println(string(entry.Key), string(entry.Value))
			}
		}
		return nil
	}); err != nil {
		log.Fatal(err)
}

Prefix search scans

To iterate over a key prefix with search by regular expression on a second part of key without prefix, we can use PrefixSearchScan function, and the parameters offsetNum, limitNum constrain the number of entries returned :

if err := db.View(
	func(tx *nutsdb.Tx) error {
		prefix := []byte("user_")
		reg := "username"
		bucket := "user_list"
		// Constrain 100 entries returned 
		if entries, _, err := tx.PrefixSearchScan(bucket, prefix, reg, 25, 100); err != nil {
			return err
		} else {
			for _, entry := range entries {
				fmt.Println(string(entry.Key), string(entry.Value))
			}
		}
		return nil
	}); err != nil {
		log.Fatal(err)
}

Range scans

To scan over a range, we can use RangeScan function. For example：

if err := db.View(
	func(tx *nutsdb.Tx) error {
		// Assume key from user_0000000 to user_9999999.
		// Query a specific user key range like this.
		start := []byte("user_0010001")
		end := []byte("user_0010010")
		bucket := "user_list"
		if entries, err := tx.RangeScan(bucket, start, end); err != nil {
			return err
		} else {
			for _, entry := range entries {
				fmt.Println(string(entry.Key), string(entry.Value))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

Get all

To scan all keys and values of the bucket stored, we can use GetAll function. For example:

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "user_list"
		entries, err := tx.GetAll(bucket)
		if err != nil {
			return err
		}

		for _, entry := range entries {
			fmt.Println(string(entry.Key),string(entry.Value))
		}

		return nil
	}); err != nil {
	log.Println(err)
}

Merge Operation

NutsDB supports merge operation. you can use db.Merge() function removes dirty data and reduce data redundancy. Call this function from a read-write transaction. It will effect other write request. So you can execute it at the appropriate time.

err := db.Merge()
if err != nil {
    ...
}

Notice: the HintBPTSparseIdxMode mode does not support the merge operation of the current version.

Database backup

NutsDB is easy to backup. You can use the db.Backup() function at given dir, call this function from a read-only transaction, it will perform a hot backup and not block your other database reads and writes.

err = db.Backup(dir)
if err != nil {
   ...
}

Using other data structures

The syntax here is modeled after Redis commands

List

RPush

Inserts the values at the tail of the list stored in the bucket at given bucket,key and values.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "bucketForList"
		key := []byte("myList")
		val := []byte("val1")
		return tx.RPush(bucket, key, val)
	}); err != nil {
	log.Fatal(err)
}

LPush

Inserts the values at the head of the list stored in the bucket at given bucket,key and values.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	        bucket := "bucketForList"
		key := []byte("myList")
		val := []byte("val2")
		return tx.LPush(bucket, key, val)
	}); err != nil {
	log.Fatal(err)
}

LPop

Removes and returns the first element of the list stored in the bucket at given bucket and key.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	        bucket := "bucketForList"
		key := []byte("myList")
		if item, err := tx.LPop(bucket, key); err != nil {
			return err
		} else {
			fmt.Println("LPop item:", string(item))
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

LPeek

Returns the first element of the list stored in the bucket at given bucket and key.

if err := db.View(
	func(tx *nutsdb.Tx) error {
	        bucket := "bucketForList"
		key := []byte("myList")
		if item, err := tx.LPeek(bucket, key); err != nil {
			return err
		} else {
			fmt.Println("LPeek item:", string(item)) //val11
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

RPop

Removes and returns the last element of the list stored in the bucket at given bucket and key.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	        bucket := "bucketForList"
		key := []byte("myList")
		if item, err := tx.RPop(bucket, key); err != nil {
			return err
		} else {
			fmt.Println("RPop item:", string(item))
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

RPeek

Returns the last element of the list stored in the bucket at given bucket and key.

if err := db.View(
	func(tx *nutsdb.Tx) error {
	        bucket := "bucketForList"
		key := []byte("myList")
		if item, err := tx.RPeek(bucket, key); err != nil {
			return err
		} else {
			fmt.Println("RPeek item:", string(item))
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

LRange

Returns the specified elements of the list stored in the bucket at given bucket,key, start and end. The offsets start and stop are zero-based indexes 0 being the first element of the list (the head of the list), 1 being the next element and so on. Start and end can also be negative numbers indicating offsets from the end of the list, where -1 is the last element of the list, -2 the penultimate element and so on.

if err := db.View(
	func(tx *nutsdb.Tx) error {
	        bucket := "bucketForList"
		key := []byte("myList")
		if items, err := tx.LRange(bucket, key, 0, -1); err != nil {
			return err
		} else {
			//fmt.Println(items)
			for _, item := range items {
				fmt.Println(string(item))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

LRem

Note: This feature can be used starting from v0.6.0

Removes the first count occurrences of elements equal to value from the list stored in the bucket at given bucket,key,count. The count argument influences the operation in the following ways:

• count > 0: Remove elements equal to value moving from head to tail.
• count < 0: Remove elements equal to value moving from tail to head.
• count = 0: Remove all elements equal to value.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	        bucket := "bucketForList"
		key := []byte("myList")
		return tx.LRem(bucket, key, 1, []byte("value11))
	}); err != nil {
	log.Fatal(err)
}

LSet

Sets the list element at index to value.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	        bucket := "bucketForList"
		key := []byte("myList")
		if err := tx.LSet(bucket, key, 0, []byte("val11")); err != nil {
			return err
		} else {
			fmt.Println("LSet ok, index 0 item value => val11")
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

Ltrim

Trims an existing list so that it will contain only the specified range of elements specified. the offsets start and stop are zero-based indexes 0 being the first element of the list (the head of the list), 1 being the next element and so on.Start and end can also be negative numbers indicating offsets from the end of the list, where -1 is the last element of the list, -2 the penultimate element and so on.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	        bucket := "bucketForList"
		key := []byte("myList")
		return tx.LTrim(bucket, key, 0, 1)
	}); err != nil {
	log.Fatal(err)
}

LSize

Returns the size of key in the bucket in the bucket at given bucket and key.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	        bucket := "bucketForList"
		key := []byte("myList")
		if size,err := tx.LSize(bucket, key); err != nil {
			return err
		} else {
			fmt.Println("myList size is ",size)
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

Set

SAdd

Adds the specified members to the set stored int the bucket at given bucket,key and items.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
	        bucket := "bucketForSet"
		key := []byte("mySet")
		return tx.SAdd(bucket, key, []byte("a"), []byte("b"), []byte("c"))
	}); err != nil {
	log.Fatal(err)
}

SAreMembers

Returns if the specified members are the member of the set int the bucket at given bucket,key and items.

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "bucketForSet"
		key := []byte("mySet")
		if ok, err := tx.SAreMembers(bucket, key, []byte("a"), []byte("b"), []byte("c")); err != nil {
			return err
		} else {
			fmt.Println("SAreMembers:", ok)
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SCard

Returns the set cardinality (number of elements) of the set stored in the bucket at given bucket and key.

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "bucketForSet"
		key := []byte("mySet")
		if num, err := tx.SCard(bucket, key); err != nil {
			return err
		} else {
			fmt.Println("SCard:", num)
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SDiffByOneBucket

Returns the members of the set resulting from the difference between the first set and all the successive sets in one bucket.

key1 := []byte("mySet1")
key2 := []byte("mySet2")
bucket := "bucketForSet"

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket, key1, []byte("a"), []byte("b"), []byte("c"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket, key2, []byte("c"), []byte("d"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if items, err := tx.SDiffByOneBucket(bucket, key1, key2); err != nil {
			return err
		} else {
			fmt.Println("SDiffByOneBucket:", items)
			for _, item := range items {
				fmt.Println("item", string(item))
			}
			//item a
			//item b
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SDiffByTwoBuckets

Returns the members of the set resulting from the difference between the first set and all the successive sets in two buckets.

bucket1 := "bucket1"
key1 := []byte("mySet1")

bucket2 := "bucket2"
key2 := []byte("mySet2")

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket1, key1, []byte("a"), []byte("b"), []byte("c"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket2, key2, []byte("c"), []byte("d"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if items, err := tx.SDiffByTwoBuckets(bucket1, key1, bucket2, key2); err != nil {
			return err
		} else {
			fmt.Println("SDiffByTwoBuckets:", items)
			for _, item := range items {
				fmt.Println("item", string(item))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SHasKey

Returns if the set in the bucket at given bucket and key.

bucket := "bucketForSet"

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if ok, err := tx.SHasKey(bucket, []byte("mySet")); err != nil {
			return err
		} else {
			fmt.Println("SHasKey", ok)
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SIsMember

Returns if member is a member of the set stored int the bucket at given bucket,key and item.

bucket := "bucketForSet"

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if ok, err := tx.SIsMember(bucket, []byte("mySet"), []byte("a")); err != nil {
			return err
		} else {
			fmt.Println("SIsMember", ok)
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SMembers

Returns all the members of the set value stored int the bucket at given bucket and key.

bucket := "bucketForSet"

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if items, err := tx.SMembers(bucket, []byte("mySet")); err != nil {
			return err
		} else {
			fmt.Println("SMembers", items)
			for _, item := range items {
				fmt.Println("item", string(item))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SMoveByOneBucket

Moves member from the set at source to the set at destination in one bucket.

bucket3 := "bucket3"

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return SAdd(bucket3, []byte("mySet1"), []byte("a"), []byte("b"), []byte("c"))
	}); err != nil {
	log.Fatal(err)
}
if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket3, []byte("mySet2"), []byte("c"), []byte("d"), []byte("e"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		if ok, err := tx.SMoveByOneBucket(bucket3, []byte("mySet1"), []byte("mySet2"), []byte("a")); err != nil {
			return err
		} else {
			fmt.Println("SMoveByOneBucket", ok)
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if items, err := tx.SMembers(bucket3, []byte("mySet1")); err != nil {
			return err
		} else {
			fmt.Println("after SMoveByOneBucket bucket3 mySet1 SMembers", items)
			for _, item := range items {
				fmt.Println("item", string(item))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if items, err := tx.SMembers(bucket3, []byte("mySet2")); err != nil {
			return err
		} else {
			fmt.Println("after SMoveByOneBucket bucket3 mySet2 SMembers", items)
			for _, item := range items {
				fmt.Println("item", string(item))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SMoveByTwoBuckets

Moves member from the set at source to the set at destination in two buckets.

bucket4 := "bucket4"
bucket5 := "bucket5"
if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket4, []byte("mySet1"), []byte("a"), []byte("b"), []byte("c"))
	}); err != nil {
	log.Fatal(err)
}
if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket5, []byte("mySet2"), []byte("c"), []byte("d"), []byte("e"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		if ok, err := tx.SMoveByTwoBuckets(bucket4, []byte("mySet1"), bucket5, []byte("mySet2"), []byte("a")); err != nil {
			return err
		} else {
			fmt.Println("SMoveByTwoBuckets", ok)
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if items, err := tx.SMembers(bucket4, []byte("mySet1")); err != nil {
			return err
		} else {
			fmt.Println("after SMoveByTwoBuckets bucket4 mySet1 SMembers", items)
			for _, item := range items {
				fmt.Println("item", string(item))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if items, err := tx.SMembers(bucket5, []byte("mySet2")); err != nil {
			return err
		} else {
			fmt.Println("after SMoveByTwoBuckets bucket5 mySet2 SMembers", items)
			for _, item := range items {
				fmt.Println("item", string(item))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SPop

Removes and returns one or more random elements from the set value store in the bucket at given bucket and key.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		key := []byte("mySet")
		if item, err := tx.SPop(bucket, key); err != nil {
			return err
		} else {
			fmt.Println("SPop item from mySet:", string(item))
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SRem

Removes the specified members from the set stored int the bucket at given bucket,key and items.

bucket6:="bucket6"
if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket6, []byte("mySet"), []byte("a"), []byte("b"), []byte("c"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		if err := tx.SRem(bucket6, []byte("mySet"), []byte("a")); err != nil {
			return err
		} else {
			fmt.Println("SRem ok")
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if items, err := tx.SMembers(bucket6, []byte("mySet")); err != nil {
			return err
		} else {
			fmt.Println("SMembers items:", items)
			for _, item := range items {
				fmt.Println("item:", string(item))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SUnionByOneBucket

The members of the set resulting from the union of all the given sets in one bucket.

bucket7 := "bucket1"
key1 := []byte("mySet1")
key2 := []byte("mySet2")

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket7, key1, []byte("a"), []byte("b"), []byte("c"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket7, key2, []byte("c"), []byte("d"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if items, err := tx.SUnionByOneBucket(bucket7, key1, key2); err != nil {
			return err
		} else {
			fmt.Println("SUnionByOneBucket:", items)
			for _, item := range items {
				fmt.Println("item", string(item))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

SUnionByTwoBuckets

The members of the set resulting from the union of all the given sets in two buckets.

bucket8 := "bucket1"
key1 := []byte("mySet1")

bucket9 := "bucket2"
key2 := []byte("mySet2")

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket8, key1, []byte("a"), []byte("b"), []byte("c"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		return tx.SAdd(bucket9, key2, []byte("c"), []byte("d"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		if items, err := tx.SUnionByTwoBuckets(bucket8, key1, bucket9, key2); err != nil {
			return err
		} else {
			fmt.Println("SUnionByTwoBucket:", items)
			for _, item := range items {
				fmt.Println("item", string(item))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

Sorted Set

ZAdd

Adds the specified member with the specified score and the specified value to the sorted set stored at bucket.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet1"
		key := []byte("key1")
		return tx.ZAdd(bucket, key, 1, []byte("val1"))
	}); err != nil {
	log.Fatal(err)
}

ZCard

Returns the sorted set cardinality (number of elements) of the sorted set stored at bucket.

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet1"
		if num, err := tx.ZCard(bucket); err != nil {
			return err
		} else {
			fmt.Println("ZCard num", num)
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZCount

Returns the number of elements in the sorted set at bucket with a score between min and max and opts.

Opts includes the following parameters:

• Limit int // limit the max nodes to return
• ExcludeStart bool // exclude start value, so it search in interval (start, end] or (start, end)
• ExcludeEnd bool // exclude end value, so it search in interval [start, end) or (start, end)

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet1"
		if num, err := tx.ZCount(bucket, 0, 1, nil); err != nil {
			return err
		} else {
			fmt.Println("ZCount num", num)
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZGetByKey

Returns node in the bucket at given bucket and key.

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet1"
		key := []byte("key2")
		if node, err := tx.ZGetByKey(bucket, key); err != nil {
			return err
		} else {
			fmt.Println("ZGetByKey key2 val:", string(node.Value))
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZMembers

Returns all the members of the set value stored at bucket.

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet1"
		if nodes, err := tx.ZMembers(bucket); err != nil {
			return err
		} else {
			fmt.Println("ZMembers:", nodes)

			for _, node := range nodes {
				fmt.Println("member:", node.Key(), string(node.Value))
			}
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZPeekMax

Returns the member with the highest score in the sorted set stored at bucket.

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet1"
		if node, err := tx.ZPeekMax(bucket); err != nil {
			return err
		} else {
			fmt.Println("ZPeekMax:", string(node.Value))
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZPeekMin

Returns the member with lowest score in the sorted set stored at bucket.

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet1"
		if node, err := tx.ZPeekMin(bucket); err != nil {
			return err
		} else {
			fmt.Println("ZPeekMin:", string(node.Value))
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZPopMax

Removes and returns the member with the highest score in the sorted set stored at bucket.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet1"
		if node, err := tx.ZPopMax(bucket); err != nil {
			return err
		} else {
			fmt.Println("ZPopMax:", string(node.Value)) //val3
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZPopMin

Removes and returns the member with the lowest score in the sorted set stored at bucket.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet1"
		if node, err := tx.ZPopMin(bucket); err != nil {
			return err
		} else {
			fmt.Println("ZPopMin:", string(node.Value)) //val1
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZRangeByRank

Returns all the elements in the sorted set in one bucket at bucket and key with a rank between start and end (including elements with rank equal to start or end).

// ZAdd add items
if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet2"
		key1 := []byte("key1")
		return tx.ZAdd(bucket, key1, 1, []byte("val1"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet2"
		key2 := []byte("key2")
		return tx.ZAdd(bucket, key2, 2, []byte("val2"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet2"
		key3 := []byte("key3")
		return tx.ZAdd(bucket, key3, 3, []byte("val3"))
	}); err != nil {
	log.Fatal(err)
}

// ZRangeByRank
if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet2"
		if nodes, err := tx.ZRangeByRank(bucket, 1, 2); err != nil {
			return err
		} else {
			fmt.Println("ZRangeByRank nodes :", nodes)
			for _, node := range nodes {
				fmt.Println("item:", node.Key(), node.Score())
			}
			
			//item: key1 1
			//item: key2 2
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZRangeByScore

Returns all the elements in the sorted set at key with a score between min and max. And the parameter Opts is the same as ZCount's.

// ZAdd
if err := db.Update(
		func(tx *nutsdb.Tx) error {
			bucket := "myZSet3"
			key1 := []byte("key1")
			return tx.ZAdd(bucket, key1, 70, []byte("val1"))
		}); err != nil {
		log.Fatal(err)
	}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet3"
		key2 := []byte("key2")
		return tx.ZAdd(bucket, key2, 90, []byte("val2"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet3"
		key3 := []byte("key3")
		return tx.ZAdd(bucket, key3, 86, []byte("val3"))
	}); err != nil {
	log.Fatal(err)
}

// ZRangeByScore
if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet3"
		if nodes, err := tx.ZRangeByScore(bucket, 80, 100,nil); err != nil {
			return err
		} else {
			fmt.Println("ZRangeByScore nodes :", nodes)
			for _, node := range nodes {
				fmt.Println("item:", node.Key(), node.Score())
			}
			//item: key3 86
			//item: key2 90
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}	

ZRank

Returns the rank of member in the sorted set stored in the bucket at given bucket and key, with the scores ordered from low to high.

// ZAdd
if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet4"
		key1 := []byte("key1")
		return tx.ZAdd(bucket, key1, 70, []byte("val1"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet4"
		key2 := []byte("key2")
		return tx.ZAdd(bucket, key2, 90, []byte("val2"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet4"
		key3 := []byte("key3")
		return tx.ZAdd(bucket, key3, 86, []byte("val3"))
	}); err != nil {
	log.Fatal(err)
}

// ZRank
if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet4"
		key1 := []byte("key1")
		if rank, err := tx.ZRank(bucket, key1); err != nil {
			return err
		} else {
			fmt.Println("key1 ZRank :", rank) // key1 ZRank : 1
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZRevRank

Returns the rank of member in the sorted set stored in the bucket at given bucket and key,with the scores ordered from high to low.

// ZAdd
if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet8"
		key1 := []byte("key1")
		return tx.ZAdd(bucket, key1, 10, []byte("val1"))
	}); err != nil {
	log.Fatal(err)
}
if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet8"
		key2 := []byte("key2")
		return tx.ZAdd(bucket, key2, 20, []byte("val2"))
	}); err != nil {
	log.Fatal(err)
}
if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet8"
		key3 := []byte("key3")
		return tx.ZAdd(bucket, key3, 30, []byte("val3"))
	}); err != nil {
	log.Fatal(err)
}

// ZRevRank
if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet8"
		if rank, err := tx.ZRevRank(bucket, []byte("key3")); err != nil {
			return err
		} else {
			fmt.Println("ZRevRank key1 rank:", rank) //ZRevRank key3 rank: 1
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZRem

Removes the specified members from the sorted set stored in one bucket at given bucket and key.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet5"
		key1 := []byte("key1")
		return tx.ZAdd(bucket, key1, 10, []byte("val1"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet5"
		key2 := []byte("key2")
		return tx.ZAdd(bucket, key2, 20, []byte("val2"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet5"
		if nodes,err := tx.ZMembers(bucket); err != nil {
			return err
		} else {
			fmt.Println("before ZRem key1, ZMembers nodes",nodes)
			for _,node:=range nodes {
				fmt.Println("item:",node.Key(),node.Score())
			}
		}
		// before ZRem key1, ZMembers nodes map[key1:0xc00008cfa0 key2:0xc00008d090]
		// item: key1 10
		// item: key2 20
		return nil
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet5"
		if err := tx.ZRem(bucket, "key1"); err != nil {
			return err
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet5"
		if nodes,err := tx.ZMembers(bucket); err != nil {
			return err
		} else {
			fmt.Println("after ZRem key1, ZMembers nodes",nodes)
			for _,node:=range nodes {
				fmt.Println("item:",node.Key(),node.Score())
			}
			// after ZRem key1, ZMembers nodes map[key2:0xc00008d090]
			// item: key2 20
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZRemRangeByRank

Removes all elements in the sorted set stored in one bucket at given bucket with rank between start and end. The rank is 1-based integer. Rank 1 means the first node; Rank -1 means the last node.

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet6"
		key1 := []byte("key1")
		return tx.ZAdd(bucket, key1, 10, []byte("val1"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet6"
		key2 := []byte("key2")
		return tx.ZAdd(bucket, key2, 20, []byte("val2"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet6"
		key3 := []byte("key3")
		return tx.ZAdd(bucket, key3, 30, []byte("val2"))
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet6"
		if nodes,err := tx.ZMembers(bucket); err != nil {
			return err
		} else {
			fmt.Println("before ZRemRangeByRank, ZMembers nodes",nodes)
			for _,node:=range nodes {
				fmt.Println("item:",node.Key(),node.Score())
			}
			// before ZRemRangeByRank, ZMembers nodes map[key3:0xc00008d450 key1:0xc00008d270 key2:0xc00008d360]
			// item: key1 10
			// item: key2 20
			// item: key3 30
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

if err := db.Update(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet6"
		if err := tx.ZRemRangeByRank(bucket, 1,2); err != nil {
			return err
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet6"
		if nodes,err := tx.ZMembers(bucket); err != nil {
			return err
		} else {
			fmt.Println("after ZRemRangeByRank, ZMembers nodes",nodes)
			for _,node:=range nodes {
				fmt.Println("item:",node.Key(),node.Score())
			}
			// after ZRemRangeByRank, ZMembers nodes map[key3:0xc00008d450]
			// item: key3 30
			// key1 ZScore 10
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

ZScore

Returns the score of member in the sorted set in the bucket at given bucket and key.

if err := db.View(
	func(tx *nutsdb.Tx) error {
		bucket := "myZSet7"
		if score,err := tx.ZScore(bucket, []byte("key1")); err != nil {
			return err
		} else {
			fmt.Println("ZScore key1 score:",score)
		}
		return nil
	}); err != nil {
	log.Fatal(err)
}

Comparison with other databases

BoltDB

BoltDB is similar to NutsDB, both use B+tree and support transaction. However, Bolt uses a B+tree internally and only a single file, and NutsDB is based on bitcask model with multiple log files. NutsDB supports TTL and many data structures, but BoltDB not supports them .

LevelDB, RocksDB

LevelDB and RocksDB are based on a log-structured merge-tree (LSM tree).An LSM tree optimizes random writes by using a write ahead log and multi-tiered, sorted files called SSTables. LevelDB does not have transactions. It supports batch writing of key/values pairs and it supports read snapshots but it will not give you the ability to do a compare-and-swap operation safely. NutsDB supports many data structures, but they not support them.

Badger

Badger is based in LSM tree with value log. It designed for SSDs. It also supports transaction and TTL. But in my benchmark its write performance is not as good as i thought. In addition, NutsDB supports data structures such as list、set、sorted set, but Badger not supports them.

Benchmarks

Tested kvstore

Selected kvstore which is embedded, persistence and support transactions.

• BadgerDB (master branch with default options)
• BoltDB (master branch with default options)
• NutsDB (master branch with default options or custom options)

Benchmark System:

• Go Version : go1.11.4 darwin/amd64
• OS: Mac OS X 10.13.6
• Architecture: x86_64
• 16 GB 2133 MHz LPDDR3
• CPU: 3.1 GHz Intel Core i7

Benchmark results:

badger 2019/03/11 18:06:05 INFO: All 0 tables opened in 0s
goos: darwin
goarch: amd64
pkg: github.com/xujiajun/kvstore-bench
BenchmarkBadgerDBPutValue64B-8    	   10000	    112382 ns/op	    2374 B/op	      74 allocs/op
BenchmarkBadgerDBPutValue128B-8   	   20000	     94110 ns/op	    2503 B/op	      74 allocs/op
BenchmarkBadgerDBPutValue256B-8   	   20000	     93480 ns/op	    2759 B/op	      74 allocs/op
BenchmarkBadgerDBPutValue512B-8   	   10000	    101407 ns/op	    3271 B/op	      74 allocs/op
BenchmarkBadgerDBGet-8            	 1000000	      1552 ns/op	     416 B/op	       9 allocs/op
BenchmarkBoltDBPutValue64B-8      	   10000	    203128 ns/op	   21231 B/op	      62 allocs/op
BenchmarkBoltDBPutValue128B-8     	    5000	    229568 ns/op	   13716 B/op	      64 allocs/op
BenchmarkBoltDBPutValue256B-8     	   10000	    196513 ns/op	   17974 B/op	      64 allocs/op
BenchmarkBoltDBPutValue512B-8     	   10000	    199805 ns/op	   17064 B/op	      64 allocs/op
BenchmarkBoltDBGet-8              	 1000000	      1122 ns/op	     592 B/op	      10 allocs/op
BenchmarkNutsDBPutValue64B-8      	   30000	     53614 ns/op	     626 B/op	      14 allocs/op
BenchmarkNutsDBPutValue128B-8     	   30000	     51998 ns/op	     664 B/op	      13 allocs/op
BenchmarkNutsDBPutValue256B-8     	   30000	     53958 ns/op	     920 B/op	      13 allocs/op
BenchmarkNutsDBPutValue512B-8     	   30000	     55787 ns/op	    1432 B/op	      13 allocs/op
BenchmarkNutsDBGet-8              	 2000000	       661 ns/op	      88 B/op	       3 allocs/op
BenchmarkNutsDBGetByHintKey-8     	   50000	     27255 ns/op	     840 B/op	      16 allocs/op
PASS
ok  	github.com/xujiajun/kvstore-bench	83.856s

Conclusions:

Put(write) Performance:

NutsDB is fastest. NutsDB is 2-5x faster than BoltDB, 0.5-2x faster than BadgerDB. And BadgerDB is 1-3x faster than BoltDB.

Get(read) Performance:

All are fast. And NutsDB is 1x faster than others. And NutsDB reads with HintKey option is much slower than its default option way.

the benchmark code can be found in the gokvstore-bench repo.

Caveats & Limitations

Index mode

From the version v0.3.0, NutsDB supports two modes about entry index: HintKeyValAndRAMIdxMode and HintKeyAndRAMIdxMode. From the version v0.5.0, NutsDB supports HintBPTSparseIdxMode mode.

The default mode use HintKeyValAndRAMIdxMode, entries are indexed base on RAM, so its read/write performance is fast. but can’t handle databases much larger than the available physical RAM. If you set the HintKeyAndRAMIdxMode mode, HintIndex will not cache the value of the entry. Its write performance is also fast. To retrieve a key by seeking to offset relative to the start of the data file, so its read performance more slowly that RAM way, but it can save memory. The mode HintBPTSparseIdxMode is based b+ tree sparse index, this mode saves memory very much (1 billion data only uses about 80MB of memory). And other data structures such as list, set, sorted set only supported with mode HintKeyValAndRAMIdxMode. It cannot switch back and forth between modes because the index structure is different.

NutsDB will truncate data file if the active file is larger than SegmentSize, so the size of an entry can not be set larger than SegmentSize , defalut SegmentSize is 8MB, you can set it(opt.SegmentSize) as option before DB opening. Once set, it cannot be changed.

Support OS

NutsDB currently works on Mac OS, Linux and Windows.

About merge operation

The HintBPTSparseIdxMode mode does not support the merge operation of the current version.

About transactions

Recommend use the latest version.

Contact

• xujiajun

Contributing

See CONTRIBUTING for details on submitting patches and the contribution workflow.

Acknowledgements

This package is inspired by the following:

• Bitcask-intro
• BoltDB
• BuntDB
• Redis
• Sorted Set

License

The NutsDB is open-sourced software licensed under the Apache 2.0 license.