I use a couple of venues to generate new material, new talking material every year.  Hotsos symposium has always been one, Oracle University Seminars have been another.  Oracle OpenWorld is definitely one as well.

Well, believe it or not, it is time for me to submit my abstract for Oracle OpenWorld already - even though it is not until September.  So, it is time to come up with yet another topic.  I've been known to do a "top ten" style presentation for a while at OpenWorld - the last few years have been top tens on 9i Release 2, 10g Releases 1 and 2 and most recently - 11g Release 1.  Intermixed with that was a database 'worst practices' presentation as well.

So, now - I'm looking for some fresh ideas.  A topic that can be well covered in an hour and would have broad general appeal.  In reality - if I get a couple of good ideas, I'll probably generate the material for all of them but just pick one (maybe two) for presentation at OpenWorld. 

So, I invite you to submit an idea - doesn't have to be fully fleshed out - just a concept, something you would find useful at a venue such as OpenWorld (or a presentation to a user group - as this is where this material will be used time and time again...)

Thanks in advance for any ideas!

Welcome, readers, to the 92^nd Log Buffer, the weekly review of database blogs.

Brian “Krow” Aker started an interesting blog-thread with his post, The Death of Read Replication, the crux of which is that object caches, such as memcached, make the DBMS itself a little less central, particularly in “Web 2.0″ scenarios. “What does this mean? Less database servers. Bringing down your load means you push off the load to another tier. . . . Why do I need to go through MySQL at all… unless I just want it as a backup or for ad-hoc reporting?”

Ronald Bradford responds with an overview of the MySQL-plus-replication scene. Farhan Mashraqi concurred with Brian’s post, while Arjen also agrees, adding, “I’m not sure the new memory based MySQL storage engines coming out are so relevant, they might be fixing the wrong thing in the wrong place.”

Ronald (who, by the way, is on-deck for a his third Log Buffer on the 25^th) also surveys both the storage-engine stuff to be had at the MySQL Conference, and the prevalence lately of talk about Kickfire in MySQL blogs, something also mentioned by Peter Zaitsev on the MySQL Performance Blog.

Peter has another question on his mind: should you have your swap file enabled while running MySQL? He wants to hear your approach to this matter, having himself experienced variable results. Lots of responses already.

Here on the Pythian Group Blog, Paul Moen posted about a situation in which SHOW SLAVE STATUS lies.

Moving into Oracle stuff, our Alex Gorbachev also pointed out something that doesn’t quite work: the ASMCMD cp command in ASM 11g. He sure gives it a try, but finally concludes: “I couldn’t make the cp command work even a single time.” Except maybe on datafiles.

(more…)

作者：Fenng 发布在 dbanotes.net.

我算得上是 Movable Type 的忠实用户。这两年来，WordPress 快成为 独立 Blog 的标准配置了，MT 可以说每况愈下。自从升级到 MT4 之后，明显感觉 Blog 性能糟糕了很多，每一次发文章都要与 HTTP 500 错误战斗半天才能成功，读者留言也非常费劲，有段时间非常灰心丧气，都不想继续写下去了。如果说 MT 3 之前是与 SPAM 战斗，到了 MT 4 就完全是折腾性能问题了。虽然 MT 4 之后的每次小版本发布都宣称性能上有很大改进，实际表现总是欠佳。

一些常规的 MT 优化技巧 其实作用都不大，我还把 MT 用的数据库索引分析了一遍，删掉了好几个价值不大的索引，虽然不是无用功，但也解决不了问题，因为瓶颈根本不在 DB 这里。大多数 Movable Type 用户的应用场景是在 Dreamhost 上，在这个环境里要想找到应用的瓶颈还真的是个技术活。缺乏 Profiling 工具不说，如果调用某个命令消耗了资源稍微过份一点，立刻被系统 Kill 掉。

其实 MT 的设计思想还是很不错的，大部分页面都 Build 成静态页面，减少对数据库的压力。一般来说，这比 WordPress 频繁调用数据库的方式(当然 WP 是可以起用 Cache 的)，应该在性能上有更好的表现才是。问题是 Dreamhost 对每个用户提供的硬件资源其实是很差的，尤其是磁盘 I/O 。我的所有页面加起来也不过几千个，如果只是访问静态页面或是数据库其实问题都不大的，麻烦出现在构建页面或者是遇到留言行为的时候，MT 会调用若干个磁盘上的模版文件，加上 Perl 本身的系统开销，消耗的资源就到了 Dreamhost 的极限，然后进程就被无情的 Kill 掉了。估计是用户群抱怨太多，从 MTOS 4.1.1 版本开始总算有了一个性能框架 专门用以分析性能问题(并且号召用户提供性能数据以便协助解决性能问题)。启用了一段时间后，收集到的 Log 响应时间典型是这样的：

MT::Template::build[Comment Response]=0.60070
MT::App::Comments::run=28.24320
Total=28.85665

或许在 I/O 不错的系统上，表现会好吧。Dreamhost 上文件都是放置在 NFS 上的，而且众多用户共用一个系统，只能将就一下。

今天看到 4.15 的改进中，已经可以 Cache Template Module 了。甘愿作小白鼠，升级。盼望这次能有效解决特定应用场景的性能问题。不过开启了性能 Log 看了半天，效果并非很显著，

--EOF--

相关文章|Related Articles

• MT安装简记
• 面向 Google 的 Movable Type 优化
• Related Entries Plugin too!
• MT 'nofollow' plugin ,联手抵制SPAM留言

评论数量(1)|Add Comments

本文网址：http://www.dbanotes.net/sitelog/movable_type_performance.html
最近作者还说了什么? Follow Twitter / Fenng

RAC的clusterware和oracle db软件都安装好了,今天在准备做ASM的时候遇到问题,配置好了ASM initfile,启动
instance的时候报错

oracle@hack3 dbs]$ sqlplus “/ as sysdba”
SQL*Plus: Release 10.2.0.1.0 - Production on Fri Apr 11 15:11:15 2008
Copyright (c) 1982, 2005, Oracle.  All rights reserved.
Connected to an idle instance.
SQL> startup;
ORA-27504: IPC error creating OSD context
ORA-27300: OS system dependent operation:unsupp_mtu failed with status: 0
ORA-27301: OS failure message: Error 0
ORA-27302: failure occurred at: skgxpvaddr10
ORA-27303: additional information: requested interface eth1 mtu not supported. Check output from ifconfig command
SQL>     exit

alert日志就记录下一点信息:
USER: terminating instance due to error 27504
Instance terminated by USER, pid = 13685

在metalink和google上找了一通,不知所云,从最后的错误看出,eth1网卡存在问题,他是我配置的private IP,走的刀片的高速背板,
各个接点间走private都是可以ping通的,通过ifconfig也看到各个网卡的MTU都是设定的,MTU:16896,我是个网络盲,一时也找不到方法了

回家的路上和sa涛哥聊了下MTU的问题,他一听说我机器网卡的MTU值是16896,就感觉好笑,说从来没有设置这么大的,建议我弄成1500,
太大了发送网络包就会造成数据包被拆分,影响效率,丢包率也严重

后来经过调整,立杆见影
eth0      Link encap:Ethernet  HWaddr D2:94:27:00:01:15 
          inet addr:10.0.38.164  Bcast:10.0.38.255  Mask:255.255.255.0
          inet6 addr: fe80::d094:27ff:fe00:115/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:17761 errors:0 dropped:0 overruns:0 frame:0
          TX packets:832 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:2000
          RX bytes:1642243 (1.5 MiB)  TX bytes:102676 (100.2 KiB)

eth0:1    Link encap:Ethernet  HWaddr D2:94:27:00:01:15 
          inet addr:10.0.38.174  Bcast:10.0.38.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

eth1      Link encap:Ethernet  HWaddr D2:94:27:00:01:16 
          inet addr:192.168.0.3  Bcast:192.168.0.255  Mask:255.255.255.0
          inet6 addr: fe80::d094:27ff:fe00:116/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:17522 errors:0 dropped:0 overruns:0 frame:0
          TX packets:15782 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:2000
          RX bytes:1674655 (1.5 MiB)  TX bytes:3784944 (3.6 MiB)

         
SQL> startup nomount;
ASM instance started

Total System Global Area  130023424 bytes
Fixed Size                  2019032 bytes
Variable Size             102838568 bytes
ASM Cache                  25165824 bytes

看来没有不可能发生的事情