================================================================
nessDB v2.0 with Small-Splittable Tree (one of Bε-tree)
Copyright (C) 2012 BohuTANG________________
_____________________________ __ \__ __ )
__ __ \ _ \_ ___/_ ___/_ / / /_ __ |
_ / / / __/(__ )_(__ )_ /_/ /_ /_/ /
/_/ /_/\___//____/ /____/ /_____/ /_____/
================================================================
nessDB is a fast Key-Value database(embedded), which is written in ANSI C with BSD LICENSE and works in most POSIX systems without external dependencies.
nessDB is very efficient on disk-based random access, I'M SURE.
From Version 2.0, nessDB uses Small-Splittable Tree (SST), it's one of Bε-tree(ε=1/2) and Cache-oblivious write-optimized structure, for more details, please refer to documentations which in 'spec' folder.
V2.0 FEATURES
=============
a. Better performances on Random-Read/Write
b. Using Small-Splittable Tree as storage engine
c. Using fractional cascading to speed up search
d. Cache-oblivious write-optimized structure
HOW TODO BENCHMARK
==================
$make db-bench
$./db-bench <write | read > <count>
or
$./db-bench readone <key>
If you want delete all indexes/db/logs, you should do:
$make cleandb
HOW TO USE AS DB ENGINE
=======================
a. OPEN
struct nessdb *db = db_open(db-path);
b. ADD
struct slice sk, sv;
sk.len = 3;
sk.data = "key";
sv.len = 5;
sv.data = "value";
db_add(db, &sk, &sv);
c. DELETE
struct slice sk;
sk.len = 3;
sk.data = "key";
db_remove(db, &sk);
d. READ
struct slice sk, sv;
sk.len = 3;
sk.data = "key";
int ret = db_get(db, &sk, &sv);
if (ret == 1) {
printf("data is %s", sv.data);
db_free_data(sv.data);
} else
print("Not found");
e. SCAN
kv = db_scan(db, &s1, &s2, limit, &ret_c);
for (i = 0; i < ret_c; i++) {
char *k = kv[i].sk.data;
char *v = kv[i].sv.data;
... ...
/* free key&value */
db_free_data(k);
db_free_data(v);
}
/* free kv list */
if (kv)
db_free_data(kv);
f. CLOSE
db_close(db);
NOTICE:
From version 2.0, the copyright from BSD change to GPL!
Thanks for your attention!
--BohuTANG--