1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
|
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN">
<html>
<head>
<link type="text/css" rel="stylesheet" href="style.css" />
</head>
<body>
<div id="page">
<div id='header'>
<a href="index.html">
<img style="border:none" alt="Redis Documentation" src="redis.png">
</a>
</div>
<div id="pagecontent">
<div class="index">
<!-- This is a (PRE) block. Make sure it's left aligned or your toc title will be off. -->
<b>Redis_1_2_0_Changelog: Contents</b><br> <a href="#What's new in Redis 1.2">What's new in Redis 1.2</a><br> <a href="#New persistence mode: Append Only File">New persistence mode: Append Only File</a><br> <a href="#New data type: sorted sets">New data type: sorted sets</a><br> <a href="#Specialized integer objects encoding">Specialized integer objects encoding</a><br> <a href="#MSET and MSETNX">MSET and MSETNX</a><br> <a href="#Better Performances">Better Performances</a><br> <a href="#Solaris Support">Solaris Support</a><br> <a href="#Support for the new generation protocol">Support for the new generation protocol</a><br> <a href="#A few new commands about already supported data types">A few new commands about already supported data types</a><br> <a href="#Bug fixing">Bug fixing</a><br> <a href="#CHANGELOG for Redis 1.1.90">CHANGELOG for Redis 1.1.90</a>
</div>
<h1 class="wikiname">Redis_1_2_0_Changelog</h1>
<div class="summary">
</div>
<div class="narrow">
<h1><a name="What's new in Redis 1.2">What's new in Redis 1.2</a></h1><h2><a name="New persistence mode: Append Only File">New persistence mode: Append Only File</a></h2>The Append Only File is an alternative way to save your data in Redis that is fully durable! Unlike the snapshotting (default) persistence mode, where the database is saved asynchronously from time to time, the Append Only File saves every change ASAP in a text-only file that works like a journal. Redis will play back this file again at startup reloading the whole dataset back in memory. Redis Append Only File supports background Log compaction. For more info read the <a href="AppendOnlyFileHowto.html">Append Only File HOWTO</a>.<h2><a name="New data type: sorted sets">New data type: sorted sets</a></h2>Sorted sets are collections of elements (like Sets) with an associated score (in the form of a double precision floating point number). Elements in a sorted set are taken in order, so for instance to take the greatest element is an O(1) operation. Insertion and deletion is O(log(N)). Sorted sets are implemented using a dual ported data structure consisting of an hash table and a skip list. For more information please read the <a href="IntroductionToRedisDataTypes.html">Introduction To Redis Data Types</a>.<h2><a name="Specialized integer objects encoding">Specialized integer objects encoding</a></h2>Redis 1.2 will use less memory than Redis 1.0 for values in Strings, Lists or Sets elements that happen to be representable as 32 or 64 bit signed integers (it depends on your arch bits for the long C type). This is totally transparent form the point of view of the user, but will safe a lot of memory (30% less in datasets where there are many integers).<h2><a name="MSET and MSETNX">MSET and MSETNX</a></h2>That is, setting multiple keys in one command, atomically. For more information see the <a href="MsetCommand.html">MSET command</a> wiki page.<h2><a name="Better Performances">Better Performances</a></h2><ul><li> 100x times faster SAVE and BGSAVE! There was a problem in the LZF lib configuration that is now resolved. The effect is this impressive speedup. Also the saving child will no longer use 100% of CPU.</li><li> Glue output buffer and writev(). Many commands producing large outputs, like LRANGE, will now be even 10 times faster, thanks to the new output buffer gluing algorithm and the (optional) use of writev(2) syscall.</li><li> Support for epool and kqueue / kevent. 10,000 clients scalability.</li><li> Much better EXPIRE support, now it's possible to work with very large sets of keys expiring in very short time without to incur in memory problems (the new algorithm expires keys in an adaptive way, so will get more aggressive if there are a lot of expiring keys)</li></ul>
<h2><a name="Solaris Support">Solaris Support</a></h2>Redis will now compile and work on Solaris without problems. Warning: the Solaris user base is very little, so Redis running on Solaris may not be as tested and stable as it is on Linux and Mac OS X.<h2><a name="Support for the new generation protocol">Support for the new generation protocol</a></h2><ul><li> Redis is now able to accept commands in a new fully binary safe way: with the new protocol keys are binary safe, not only values, and there is no distinction between bulk commands and inline commands. This new protocol is currently used only for MSET and MSETNX but at some point it will hopefully replace the old one. See the Multi Bulk Commands section in the <a href="ProtocolSpecification.html">Redis Protocol Specification</a> for more information.</li></ul>
<h2><a name="A few new commands about already supported data types">A few new commands about already supported data types</a></h2><ul><li> <a href="SrandmemberCommand.html">SRANDMEMBER</a></li><li> The <a href="SortCommand.html">SortCommand</a> is now supprots the <b>STORE</b> and <b>GET #</b> forms, the first can be used to save sorted lists, sets or sorted sets into keys for caching. Check the manual page for more information about the <b>GET #</b> form.</li><li> The new <a href="RpoplpushCommand.html">RPOPLPUSH command</a> can do many interesting magics, and a few of this are documented in the wiki page of the command.</li></ul>
<h2><a name="Bug fixing">Bug fixing</a></h2>Of course, many bugs are now fixed, and I bet, a few others introduced: this is how software works after all, so make sure to report issues in the Redis mailing list or in the Google Code issues tracker.<br/><br/>Enjoy!
antirez<h1><a name="CHANGELOG for Redis 1.1.90">CHANGELOG for Redis 1.1.90</a></h1><ul><li> 2009-09-10 in-memory specialized object encoding. (antirez)</li><li> 2009-09-17 maxmemory fixed in 64 systems for values > 4GB. (antirez)</li><li> 2009-10-07 multi-bulk protocol implemented. (antriez)</li><li> 2009-10-16 MSET and MSETNX commands implemented (antirez)</li><li> 2009-10-21 SRANDMEMBER added (antirez)</li><li> 2009-10-23 Fixed compilation in mac os x snow leopard when compiling a 32 bit binary. (antirez)</li><li> 2009-10-23 New data type: Sorted sets and Z-commands (antirez)</li><li> 2009-10-26 Solaris fixed (Alan Harder)</li><li> 2009-10-29 Fixed Issue a number of open issues (antirez)</li><li> 2009-10-30 New persistence mode: append only file (antirez)</li><li> 2009-11-01 SORT STORE option (antirez)</li><li> 2009-11-03 redis-cli now accepts a -r (repeat) switch. (antirez)</li><li> 2009-11-04 masterauth option merged (Anthony Lauzon)</li><li> 2009-11-04 redis-test is now a better Redis citizen, testing everything against DB 9 and 10 and only if this DBs are empty. (antirez)</li><li> 2009-11-10 Implemented a much better lazy expiring algorithm for EXPIRE (antirez)</li><li> 2009-11-11 RPUSHLPOP (antirez from an idea of @ezmobius)</li><li> 2009-11-12 Merge git://github.com/ianxm/redis (Can't remmber what this implements, sorry)</li><li> 2009-11-17 multi-bulk reply support for redis-bench, LRANGE speed tests (antirez)</li><li> 2009-11-17 support for writev implemented. (Stefano Barbato)</li><li> 2009-11-19 debug mode (-D) in redis-bench (antirez)</li><li> 2009-11-21 SORT GET # implemented (antirez)</li><li> 2009-11-23 ae.c made modular, with support for epoll. (antirez)</li><li> 2009-11-26 background append log rebuilding (antirez)</li><li> 2009-11-28 Added support for kqueue. (Harish Mallipeddi)</li><li> 2009-11-29 SORT support for sorted sets (antirez, thanks to @tobi for the idea)</li></ul>
</div>
</div>
</div>
</body>
</html>
|
