rocksdict 0.2.11

On-disk Dict With RocksDB (dbm alternative)

RocksDict

Key-value storage supporting any python object

Abstract

This package enables users to store, query, and delete a large number of key-value pairs on disk.

This is especially useful when the data cannot fit into RAM. If you have hundreds of GBs or many TBs of key-value data to store and query from, this is the package for you.

Installation

This package is built for macOS (x86/arm), Windows 64/32, and Linux x86. It can be installed from pypi with pip install rocksdict.

Plans

• set, get, del
• multi get
• support string, float, int, bytes
• support other python objects through pickle
• support BigInt
• compare BigInt by value size
• keys, values, items iterator
• options, read options, write options, all options
• SstFileWriter and bulk ingest
• column families
• open as secondary
• write batch

Introduction

Below is a code example that shows how to do the following:

• Create Rdict
• Store something on disk
• Close Rdict
• Open Rdict again
• Check Rdict elements
• Iterate from Rdict
• Batch get
• Delete storage

from rocksdict import Rdict
import numpy as np
import pandas as pd

path = str("./test_dict")

# create a Rdict with default options at `path`
db = Rdict(path)

db[1.0] = 1
db[1] = 1.0
db["huge integer"] = 2343546543243564534233536434567543
db["good"] = True
db["bad"] = False
db["bytes"] = b"bytes"
db["this is a list"] = [1, 2, 3]
db["store a dict"] = {0: 1}
db[b"numpy"] = np.array([1, 2, 3])
db["a table"] = pd.DataFrame({"a": [1, 2], "b": [2, 1]})

# close Rdict
db.close()

# reopen Rdict from disk
db = Rdict(path)
assert db[1.0] == 1
assert db[1] == 1.0
assert db["huge integer"] == 2343546543243564534233536434567543
assert db["good"] == True
assert db["bad"] == False
assert db["bytes"] == b"bytes"
assert db["this is a list"] == [1, 2, 3]
assert db["store a dict"] == {0: 1}
assert np.all(db[b"numpy"] == np.array([1, 2, 3]))
assert np.all(db["a table"] == pd.DataFrame({"a": [1, 2], "b": [2, 1]}))

# iterate through all elements
for k, v in db.items():
    print(f"{k} -> {v}")

# batch get:
print(db[["good", "bad", 1.0]])
# [True, False, 1]
 
# delete Rdict from dict
db.close()
Rdict.destroy(path)

Supported types:

• key: int, float, bool, str, bytes
• value: int, float, bool, str, bytes and anything that supports pickle.

Rocksdb Options

Since the backend is implemented using rocksdb, most of rocksdb options are supported:

Example of tuning

from rocksdict import Rdict, Options, SliceTransform, PlainTableFactoryOptions
import os

def db_options():
    opt = Options()
    # create table
    opt.create_if_missing(True)
    # config to more jobs
    opt.set_max_background_jobs(os.cpu_count())
    # configure mem-table to a large value (256 MB)
    opt.set_write_buffer_size(0x10000000)
    opt.set_level_zero_file_num_compaction_trigger(4)
    # configure l0 and l1 size, let them have the same size (1 GB)
    opt.set_max_bytes_for_level_base(0x40000000)
    # 256 MB file size
    opt.set_target_file_size_base(0x10000000)
    # use a smaller compaction multiplier
    opt.set_max_bytes_for_level_multiplier(4.0)
    # use 8-byte prefix (2 ^ 64 is far enough for transaction counts)
    opt.set_prefix_extractor(SliceTransform.create_max_len_prefix(8))
    # set to plain-table for better performance
    opt.set_plain_table_factory(PlainTableFactoryOptions())
    return opt

db = Rdict(str("./some_path"), db_options())

Example of Column Families

from rocksdict import Rdict, Options, SliceTransform, PlainTableFactoryOptions
import random

path = str("tmp")
cf1_name = str("cf1")
cf2_name = str("cf2")

# set cf2 as a plain table
cf2_opt = Options()
cf2_opt.set_prefix_extractor(SliceTransform.create_max_len_prefix(8))
p_opt = PlainTableFactoryOptions()
p_opt.user_key_length = 200
cf2_opt.set_plain_table_factory(p_opt)

# create column families if missing
opt = Options() # create_if_missing=True by default
opt.create_missing_column_families(True)
db = Rdict(path, options=opt, column_families={cf1_name: Options(),
                                               cf2_name: cf2_opt})

# add column families
db_cf1 = db.get_column_family(cf1_name)
db_cf2 = db.get_column_family(cf2_name)
db_cf3 = db.create_column_family(str("cf3")) # with default Options
db_cf4 = db.create_column_family(str("cf4"), cf2_opt) # custom options

# remove column families
db.drop_column_family(str("cf3"))
db.drop_column_family(str("cf4"))
del db_cf3, db_cf4

# insert into column families
for i in range(10000):
    db_cf1[i] = i ** 2

rand_bytes = [random.randbytes(200) for _ in range(100000)]
for b in rand_bytes:
    db_cf2[b] = b

# close database
db_cf1.close()
db_cf2.close()
db.close()

# reopen db
db = Rdict(path, column_families={cf1_name: Options(),
                                  cf2_name: cf2_opt})
db_cf1 = db.get_column_family(cf1_name)
db_cf2 = db.get_column_family(cf2_name)

# check keys
count = 0
for k, v in db_cf1.items():
    assert k == count
    assert v == count ** 2
    count += 1

rand_bytes.sort()
assert list(db_cf2.keys()) == rand_bytes

# delete db
db.close()
db_cf1.close()
db_cf2.close()
Rdict.destroy(path)

Example of Bulk Ingestion By SstFileWriter

from rocksdict import Rdict, Options, SstFileWriter
import random

# generate some rand bytes
rand_bytes1 = [random.randbytes(200) for _ in range(100000)]
rand_bytes1.sort()
rand_bytes2 = [random.randbytes(200) for _ in range(100000)]
rand_bytes2.sort()

# write to file1.sst
writer = SstFileWriter()
writer.open("file1.sst")
for k, v in zip(rand_bytes1, rand_bytes1):
    writer[k] = v

writer.finish()

# write to file2.sst
writer = SstFileWriter(Options())
writer.open("file2.sst")
for k, v in zip(rand_bytes2, rand_bytes2):
    writer[k] = v

writer.finish()

# Create a new Rdict with default options
d = Rdict("tmp")
d.ingest_external_file(["file1.sst", "file2.sst"])
d.close()

# reopen, check if all key-values are there
d = Rdict("tmp")
for k in rand_bytes2 + rand_bytes1:
    assert d[k] == k

d.close()

# delete tmp
Rdict.destroy("tmp")

Contribution

This project is still in an early stage of development. People are welcome to add tests, benchmarks and new features.