TABLE

Table is an in memory key-value store library

NOTE: Table is just my toy project, it may has some bugs, so do not use Table in production

Features

• Keys and values are arbitrary byte arrays
• Data is stored sorted by key
• The basic operations are put(key,value), get(key), del(key)
• Support for persisting data to disk
• Safe to use Table in multithreaded single writer multi-reader code

Build

mkdir build && cd build
cmake .. && make -j4 && make install

TODO: Add building for unit-test

Quick Start

Opening/Closing a Table

table::Options options;
options.create_if_missing = true;
table::Table table(options, "test_table");

table::Status s = table.open();
if(!s.good()) {
    std::cerr << s.string() << std::endl;
}

table.close();

Reads/Writes

std::string key = "key";
std::string value = "value";

s = table.put(key, value);
if (!s.good()) {
    std::cerr << s.string() << std::endl;
}

s = table.get(key, &value);
if (!s.good()) {
    std::cerr << s.string() << std::endl;
} else {
    std::cout << value << std::endl;
}

Persisting data

std::string key = "key";
std::string value = "value";

s = table.put(key, value);
if (!s.good()) {
    std::cerr << s.string() << std::endl;
}

s = table.dump();
if (!s.good()) {
    std::cerr << s.string() << std::endl;
}

or set options.dump_when_close = true;

Architecture

Performance

• CPU: Intel Core i5 2.3 GHz
• CPU Cores(physical) num: 2
• CPU Cores(logical) num: 4
• CPU L2 Cache：256 KB
• CPU L3 Cache：4MB
• Key Size： 16
• Value Size：100
• GCC Optimization Level：O2

Put Performance

Put 100,000 entries into a empty Table

Times	Spend
1	178ms
2	172ms
3	155ms
4	152ms
5	146ms

Put 1,000,000 entries into a empty Table

Times	Spend
1	3094ms
2	2619ms
3	2614ms
4	2614ms
5	2631ms

Get Performance

Table has 100,000 entries, and we get 10,000 different keys from Table

Times	Spend
1	10ms
2	11ms
3	12ms
4	14ms
5	12ms

Table has 1,000,000 entries, and we get 10,000 different keys from Table

Times	Spend
1	30ms
2	35ms
3	29ms
4	29ms
5	27ms