BohuTANG

Keys:		20 bytes each
Values:		100 bytes each
Entries:	5000000
IndexSize:	324.2 MB (estimated)
DataSize:	515.0 MB (estimated)
BG SYNC:	close...
------------------------------------------------------------------------------------------------------------
nessDB:		version 1.7(Multiple && Distributable B+Tree with Level-LRU,Background IO Sync)
Date:		Mon Oct 17 22:09:40 2011
+-----------------------+---------------------------+----------------------------------+---------------------+

BohuTANG

  nessDB是一个小巧、高性能的、支持Redis协议的K-V数据库，使用标准C开发，支持Linux, *BSD, OS X and Solaris 系统，无第三方依赖。

 描述：
  * 使用Bloom filter检查数据存在性
  * 使用Level-LRU提升随机读性能，类似于InnoDB的bufferpool，大小可配置。
  * 可分布存储的B+Tree索引文件
  * Redis协议支持(PING,GET,MGET,SET,MSET,DEL,INFO)：SET相同的key即更新；DEL后面跟多key，即批量删除


  源码目录结构

BohuTANG

change:
if (buf_len > 2 * rsa->sizeof_g_packet)
           error (_(“Remote ‘g’ packet reply is too long: %s”), rs->buf); 
to:

if (buf_len > 2 * rsa->sizeof_g_packet) {
      rsa->sizeof_g_packet = buf_len ;
      for (i = 0; i < gdbarch_num_regs (gdbarch); i++)
      {
         if (rsa->regs[i].pnum == -1)

BohuTANG

BohuTANG@BohuTANG:~/github/lsm-tree$ ./bench write 100000000
Keys:		16 bytes each
Values:		80 bytes each
Entries:	100000000
IndexSize:	2384.2 MB (estimated)
DataSize:	8010.9 MB (estimated)
---------------------------------------------------------------------------------------------------
nessDB:		version 1.8(LSM-Tree storage engine)
Date:		Mon Mar 19 15:51:07 2012
CPU:		4 *  Intel(R) Core(TM) i5-2450M CPU @ 2.50GHz

BohuTANG

RANDOM-WRITE RESULTS(1000k records)
===================================
Keys:		16 bytes each
Values:		80 bytes each
Entries:	1000000
IndexSize:	23.8 MB (estimated)
DataSize:	80.1 MB (estimated)
---------------------------------------------------------------------------------------------------
nessDB:		version 1.8(LSM-Tree storage engine)
Date:		Thu Mar 29 16:49:40 2012

BohuTANG

import re 
import urllib

url = "http://arxiv.org/list/cs.DB/recent"
s=urllib.urlopen(url).read() 
ss = s.replace("\n","")
titles=re.findall('<span class="descriptor">Title:</span>(.+?)</div>', ss, re.I)
i = 0;
for t in titles: 
	print '(%d) %s' %(i, t)

BohuTANG

import re 
import urllib

url = "http://arxiv.org/list/cs.DB/recent"
s=urllib.urlopen(url).read() 
ss = s.replace("\n","")
titles=re.findall('<span class="descriptor">Title:</span>(.+?)</div>', ss, re.I)
i = 0
for t in titles: 
	print '(%d) %s' %(i, t)

BohuTANG

/*
 *  Such as num is:12345678910, 'from_big1' funcation will be wrong
 *  since the sign extension: 'cltq' instruction will full fill upper 32 bits with 0xffffffff
 */
#include <stdlib.h>
#include <stdint.h>
#include <stdio.h>
#include <inttypes.h>

void to_big(unsigned char *buf, uint64_t v)

BohuTANG

diff --git a/Makefile b/Makefile
index 5631c4d..0e1e04e 100644
--- a/Makefile
+++ b/Makefile
@@ -1,9 +1,9 @@
 CC = gcc
 
 #debug levle
-DOG = -rdynamic -DDOG
-DEBUG =        -g  -ggdb -DINFO $(DOG)

BohuTANG

/*
 * CascaDB crc32 algorithm and test
 * refer to https://code.google.com/p/leveldb/source/browse/util/crc32c.cc
 * BohuTANG
 * 2013/08/27
 */

#include <stdio.h>
#include <stdint.h>
#include <string.h>