O’Reilly’s Velocity Con, of course.

Kai Hansen, Tony Ralph, Eric Goldsmith, and Artur Bergman during, This is Your Page with Ads, a panel moderated by Steve Souders.

It turns out I’m not the only person who thinks micro-optimization of CSS files is cool.  I learned this lesson a year ago when I joined the Exceptional Performance team at Yahoo! and had it reinforced by the quality of both the presentations and the hallway conversations at the O’Reilly Velocity Conference last week.

Attending Velocity Con was fabulous.  I was especially impressed that the sessions on web performance were packed.  There were a ton of Yahoos at the conference,  Julien Lecomte from Yahoo! Search spoke about “High-performance Ajax Applications”.

“In the past few years, Ajax has become very popular because it has enabled developers to build more complex web applications. However, in the rush to push the browser to new limits, we have created a monster. “ – Julien

Julien suggested several detailed strategies and patterns that developers can use to accelerate their applications. Stoyan Stefanov, the lead developer of YSlow, and my colleague in the Exceptional Performance team, spoke about Image Optimization, including the 7 mistakes most sites are making.  He showed non-designers how to automate image optimization and reduce image bloat by as much as 30%. After attending the talk, Douglas Crockford shared some love.

“It is good to be able to point with pride at something that Yahoo does that is extremely smart. The Exceptional Performance Team is one of the things that makes me proud to be at Yahoo.” – Doug

John Allspaw from Flickr joined a panel about Surviving Success by preparing to be TechCrunched, Dugg, Slashdotted, or even “Oprahed”.  He also presented Capacity Management.

“Your process of capacity planning should be adaptive, adjustable, and include more than just system statistics. Measurement, architecture, and economics are all equally important to having your site perform. Becoming popular doesn’t have to mean being afraid your site will fall over from too much load.” – John

Adam Bechtel, the chief architect covering network, storage and systems infrastructure at Yahoo! presented “Performance Plumbing”. He believes that scale provides unique opportunities to leverage the network to improve performance.

“As your site scales, don’t overlook the performance opportunities that the plumbing creates.” – Adam

Tony Ralph who works on ad quality and performance for Yahoo! participated in a panel, This is Your Page with Ads.  He made an important point that I hadn’t really thought of before. He indicated that the ad industry and engineers measure performance in very different ways; one via monetization, the other via impact on response time.  He emphaiszed how important it is for engineers to understand both points of view, so that we can effectively measure and convey the impact of end user experience on revenue.

Kai Hansen from Google Ireland also mentioned the need to properly advocate this point of view from within our companies so that quality metrics such as keyword relevance and performance are tied to the cost of displaying ads.

I look forward to Velocity Conference 2009.  I do hope that it will focus on the front end with more talks about HTML, CSS, JavaScript, and Ajax.  These sessions were the most popular of the conference, and front-end performance is still in its infancy. Douglas Crockford expressed it very well.

“By showing the browser makers how web applications actually perform, the browser makers are now able to make effective changes to the platform. As the platform evolves, we will need new rules and new tools. There is still much to do. (Emphasis mine)” – Doug

Nicole Sullivan

Exceptional Performance Yahoo!

Sixapart的动作很快，Movable Type 4.2 Release Candidate 3偷偷摸摸的发布了。

该次发布，将Movable Type产品分成了三种版本，分别是Movable Type Open Source、Movable Type Commercial、Movable Type Community Solution。Movable Type Open Source是开源的免费版本，使用Movable Type Commercial和Movable Type Community Solution则需要根据不同的情况购买符合要求的许可证。其中Movable Type Community Solution是初次发布，可以用它来建立一个Blog社区，或与现有的论他程序进行整合（官方给出的方案是使用OpenID）。对于多数人来说，一个Movable Type Open Source真的是足够了。

好了，先介绍到这里，随即奉上简繁体的中文语言包，望大家笑纳。

dbanotes posted a photo:

      事情总是一块来的。今天因为是月底最后一天，刚好还有2个数据订正要做，正在编写脚本、同开发人员确认处理逻辑，且催得很急。不巧的是，测试db这时也出问题，load很高。没办法订正先放下来，解决测试的问题，否则影响下午的测试。

  新建ssh连接没有反应，好在此前已经有一个ssh terminal连在服务器上，top命令打下去就没有响应了，等了几分钟终于出来结果。发现top的进程列表排在最前面的是两个进程kscand和kswapd,注意到swap分配的内存为0，被使用的swap内存也为0. 因为cpu太忙，输入的命令都不可能，通知测试人员先把测试服务器的应用停了下来。为什么swap内存没有分配了，fdisk -l 发现 /dev/sdg2  是swap对应的分区，检查/etc/fstab 发现swap对应的设备是/dev/sda2，奇怪磁盘设备的编号怎么不一样了？

    通过powermt查看存储的磁盘路径，有2个lun分配给系统，lun 3有2个通道设备/dev/sda,/dev/sdc,lun 8有2个通道设备/dev/sdb,/dev/sdd,记得机器安装配置时lun分配的设备应该只有lun 3。lun 8是后加的，印象中记得以前本地硬盘的设备编号是/dev/sda。应该就是这个导致了系统的load飙高，因为没有swap，导致系统的内存有4G被用作swap，导致页换进换出，系统消耗的cpu很高。swapon /dev/sdg2,编辑/etc/fstab的swap设备，重新mount -a.

    通知测试启动环境应用，故障消失。检查db日志和v$pgastat都没有发现异常，因为应用被强制停止的，所以当时在运行的sql和等待事件都没法知道。但从现在来看应该是swap导致了这次的问题，什么原因导致swap的设别名被更改了，还真不太确定，不知道跟上次动态从存储分配一个lun过来有没有关系。

   ps：

os: redhat as3 u4   机器内存8G，4cpu，sga 1.6 gb + pga  200m .

Wishlist 2008.diff
=====================
- Nintendo Wii
   24″ LCD (Provisional: Dell UltraSharp 2408WFP)
+ MSI Wind Notebook U100

2008转瞬已过去了半载，思量一下年初的Wishlist，发现尚未有一件成为现实…… 看来，是需要调整一下了。

伴随着今年Wii的出局，再加上去年Sony PSP从Wishlist的黯然隐退，我居然再次坚守住了抵御TV Game诱惑的防线。不过千万不要因此佩服我，一个WoW已经够玩物丧志了，要是Wii再来掺和一下，那后果简直不堪设想…… = =!

24″ LCD是年初的Wishlist的一部分，最近进一步明确了具体的目标。Dell的显示器好是好，不过2408WFP不降反升的价格策略确实让人有点咬牙切齿。眼看别家的24″都降到它差不多一半的价格了，我怕到时候自己可没Dell那么沉得住气了。

华硕的Eee PC燎燃了轻便笔记本市场的火爆，也让我的“笔记本鸡肋论”得到了有力的佐证。未来，高性能HTPC + 轻便笔记本的组合将会代替沉重的大尺寸转头本重新成为时尚家居的主流。可惜身为先驱者的Eee PC却在屏幕和键盘的设计上让人不敢恭维，好在烽烟四起，诸侯纷争，消费者也因此有了更多的选择。:)

With PHP 5.3 coming up on the horizon, I'm of course looking forward to using namespaces. Let's be honest, who wants to write the following line?

when the more succinct:

could be used? (Assuming you've executed 'use Zend::Controller::Action;' somewhere earlier...)

However, while namespaces will hopefully lead to more readable code, particularly code in libraries and frameworks, PHP developers will finally need to start thinking about sane standards for abstract classes and interfaces.

What is “Float”?

Float is a CSS positioning property. If you are familiar with print design, you can think of it like an image in a layout where the text wraps around it as necessary.

In web design, an image that is floated remains a part of the flow of a web page. That means that if it changes in size or if the elements around it change, the page will automatically adjust (reflow). This differs from page elements that are absolutely positioned. Absolutely positioned page elements are removed from the flow of a web page. Absolutely positioned page elements will not affect any other page elements, whether they touch each other or not.

What are floats used for?

Besides simple examples like floating an image within a block of text, floats can be used to create entire web layouts.

Floats are also helpful for layout in smaller instances. Take for example this little area of a web page:

These types of layouts can be handled using absolute positioning inside of relative positioning as well, but floats are a more flexible. Let’s say the size of that avatar image were to change. With floats, the box could re-flow to accommodate the bigger size, whereas an absolutely positioned layout would create problems:

Problems with Floats

Oh, where to begin. Floats are fragile. They are full of issues and cross-browser quirks. Perhaps most significant is the need to clear floats in some situations. There are a couple reasons floats need to be cleared. First I’ll show you some examples of why a float would need to be cleared, then we’ll cover how to go about clearing.

Clear the float to fix the height of the parent element.

Elements which contain floated elements do not calculate their height as you might expect. In fact, if the parent element contains only floated elements, browsers will render the height at zero.

If you clear the float before the closing tag of the parent element, you can fix this.

Clear the float to start a new row.

Lets say you have a grid of floated objects.

Then let’s say you want to create a break in that grid in order to start a new row. Because, you know, it just makes sense visually for whatever reason.

Clearing only the left or right.

The above two needs for clearing call for generic clearing, that is, clearing both the left and the right of the float. Because floats can be floated left or right, you can also clear either the left or right only. This can be useful when clearing both sides would be problematic to your layout.

If both sides were cleared in the above example, the second image would have been pushed down to where the text block ended.

Different techniques for clearing floats

As with everything in CSS, there are more ways than one to do this.

Applying clear exactly where you need it.

The clear CSS property does exactly what it says on the box. The problem lies in where and how to apply an element to the page with the correct clear property.

• Apply clear: both; to the element directly after where you need the float cleared. Take the example near the top of a web page layout with a full-width header and footer and main content floated to the left with a sidebar floated to the right. In order to get that footer to show up in the right place, you MUST clear the float before it. In this simple example, you could simply apply clear: both; to the footer div itself.

   

  This technique is wonderful because it works and it adds no superfluous markup to your code. However, it breaks down in more dynamic sites sometimes. What if, for example, you ended up having to add a new page element above the footer but below the main content? You would need to remember to apply the clearing to this new element now instead of the footer. It is often easier to think in terms of where you need to clear the float instead of on which element you need to clear the float (and is easier to implement).

• Clear the float with an empty page element. <div>’s do nicely because generally you don’t have any styling applied to generic div’s (like you might a <p> element) and they don’t have any other special page functionality (like a <br />). Where ever you want to clear the float just add a div like this: <div style=”clear: both;”></div>. I find looking at inline-styling unappealing so I choose to make a class for this style and apply a “clear” class to that div instead, but that is just a personal aesthetic choice.

The empty div method HTML:

    ...part of floated element
</div>

<div class="clear"></div>

<p> ... ahhhh, now I'm clear </p>

CSS:

div.clear { clear: both; }

Overflow: auto; on the parent element

I can’t explain to you why (wish I could), but applying the CSS property overflow: auto; to the parent element will cause the height to be calculated properly. It will grow to enclose all the floated elements instead of collapsing. This can be very useful and it’s very clean, but it has some issues of course. The major one being there often isn’t a parent element it makes sense to use it on. Think of times where you would need to clear the float more than once within the same single parent, this isn’t going to help you there.

Another problem is that you may want or need to use that overflow property for other purposes. What if you want to hide the overflow on that particular div? You can’t. You’ll have to wrap the div in yet another div to accomplish that.

The clearfix hack (clearing with the :after pseduo-class)

Although aging, the positioniseverything article on using a pseduo-class to clear the float is still very functional. And pretty brilliant, if you ask me. To explain simply, it adds a bit of content after the element. This bit of content (a period or space, usually) does the clearing the float, but is hidden from view.

Here is the code, applied via a class to any element you need to do the clearing:

/* This needs to be first because FF3 is now supporting this */
.clearfix {display: inline-block;}

.clearfix:after {
    content: " ";
    display: block;
    height: 0;
    clear: both;
    font-size: 0;
    visibility: hidden;
}

/* Hides from IE-mac \*/
* html .clearfix {height: 1%;}
.clearfix {display: block;}
/* End hide from IE-mac */

The article has a warning claiming that the technique is getting old and refers to the overflow: auto; fix. I don’t share their sentiment that this technique is outdated. This is fundamentally different than the overflow fix. With the clearfix, you apply the clearing class to the element itself, not a parent element. This means you can still use it even where there isn’t a parent element that makes sense and apply it by thinking abut where the float needs to be cleared.

Updates: Thanks to John for pointing out that FF3 now supports inline-block, so that statement will need to go first in the CSS so it can get over-written by the :after pseudo-class. Thanks to Brent for pointing out the Perishable Press article suggesting using a space instead of a period and adding font-size: 0;.

Another float problem: Pushdown

I feel like this deserves special attention because of how often I am helping someone troubleshoot a problem they are having with “pushdown”. Again looking back at the example at the top of this article with the Main Content area floated left and the Sidebar floated right. This is a common structure for many blogs.

Whether it’s fixed or fluid, both of these areas will have a set width. They way floats are supposed to behave, if an element within the main content area exceeds the width the the main content area (say, you inserted an image that was too large) it will simply extend beyond and overlap anything it needs to. The way IE 6 handles this, is much different. If an element within the main content area exceeds the bounds in IE 6, it will simply push the sidebar down below, completely borking your layout.

The solution? The best solutions is just not put anything inside the main content area that will spill over. For better protection, and it if it works with your layout otherwise (no set heights!), you could set the overflow to hidden to hide any spillover. Yet another solution would be to use absolute positioning to get the sidebar over on the right. Remember though, absolute positioning removes the element from the flow of the page — just something to be aware of.

A List Apart just recently had an article on how to use “faux” absolute positioning, which an interesting read and covers a new technique for layout that has a lot of the advantages of absolute positioning while retaining flow and fighting the frailty of floats.

Quirks about floats

Another thing to remember when dealing with IE 6 is that if you apply a margin in the same direction as the float, it will double the margin.

IE 7 has a little quirk of it’s own not respecting bottom margin on children inside floated objects.

©作者：eygle 发布在 eygle.com

今天帮客户安装AIX系统上的Oracle10g数据库，使用了ASM管理存储，忽然记起Nigoo提到过一个以前没有注意到的工具kfod。

这个工具可以用于手工查询ASM磁盘的信息。其帮助信息如下：

db2:/home/oracle>kfod -help
_asm_a/llow_only_raw_disks KFOD allow only raw devices [_asm_allow_only_raw_disks=TRUE/(FALSE)]
_asm_l/ibraries ASM Libraries[_asm_libraries='lib1','lib2',...]
_asms/id ASM Instance[_asmsid=sid]
a/sm_diskstring ASM Diskstring [asm_diskstring='discoverystring', 'discoverystring' ...]
d/isks Disks to discover [disks=raw,asm,all]
g/roup Group discover [group=controlfile]
n/ohdr KFOD header suppression [nohdr=TRUE/(FALSE)]
o/p KFOD options type [OP=DISKS/GROUPS/ALL]
p/file ASM parameter file [pfile='parameterfile']
s/tatus Include disk header status [status=TRUE/(FALSE)]
v/erbose KFOD verbose errors [verbose=TRUE/(FALSE)]

查看磁盘可以使用disk参数，其输出实际上是通过权限来列举可用磁盘：

db2:/home/oracle>kfod disk=all
--------------------------------------------------------------------------------
Disk Size Path
================================================================================
1: 102795 Mb /dev/rhdisk10
2: 10240 Mb /dev/rhdisk3
3: 10240 Mb /dev/rhdisk4
4: 51200 Mb /dev/rhdisk5
5: 51200 Mb /dev/rhdisk6
6: 51200 Mb /dev/rhdisk7
7: 51200 Mb /dev/rhdisk8
8: 102400 Mb /dev/rhdisk9
--------------------------------------------------------------------------------
ORACLE_SID ORACLE_HOME
================================================================================

在我的一个Linux环境上输出如下：

[oracle@dbrac1 oracle]$ kfod disk=all
--------------------------------------------------------------------------------
Disk Size Path
================================================================================
1: 97660 Mb ORCL:VOL1
2: 195321 Mb ORCL:VOL2
--------------------------------------------------------------------------------
Disk Size Path
================================================================================
1: 250 Mb /dev/raw/raw1
2: 251 Mb /dev/raw/raw2
--------------------------------------------------------------------------------
ORACLE_SID ORACLE_HOME
================================================================================
+ASM2 /opt/oracle/product/10.2.0/db
+ASM1 /opt/oracle/product/10.2.0/db

相关文章|Related Articles

• Oracle10g OCR及Voting Disk的备份
  
• Patchset 10.2.0.4中关于ASM的Buglist
  
• DBA警世录:使用ASM应当具备充分认识
  
• Oracle10.2.0.3 RAC的VIP及实例依赖关系
  
• CRS-0184错误与CRS初始化
  

评论数量(0)|Add Comments

本文网址：http://www.eygle.com/archives/2008/06/kfodoracle_asm.html

1、之前那个说淘宝交易评价“有问题”的文章已经有了一百多个评论。（某些开口就骂的评论已被删除）
近期去评论的人都是从搜索引擎过来的，他们不是产品设计人员，没有一个真正去关注我说的信用机制本身，都抓住我给了卖家中评价说事。 对于这个博客来说这样的评论已经没有意义。（这篇文章发出的同时，关闭即关闭上篇的评论）

2、其实我上篇文章只是想说一下那些界面设计实在糟糕。同时反应一下淘宝这个的评价和星钻构成的信用机制有着一些已经根深地固的问题。
据观察，这些问题事实上已经越来越明显：

1》如果不主动给评价系统默认给好评。
大多数人主动给评价都是在两种情况下：“很好”或者“很差”。不评价基本上都是“不很好也不很差”（有过定性和定量的分析，都显示的是这个结论）。这个时候默认给好评跟交易实际情况是不符的。
短期内淘宝这么作讨好了卖家也吸引了买家，问题的威胁性也不会凸显。 但长期下来(现在已经能看出来了)整个评价系统的质量被严重“稀释”，好评没有了“质量”，一个中评就能让人感觉到地震。
导致现在绝大多数淘宝的玩家的心理已经默认中评就是差评，整个评价体系处于轻微“残废”状态。如果把好和坏反过来看（为什么要反过来，不再细说），这就相当于全国人民都认为只有红灯才得注意，黄灯是可以完全被当作绿灯的，长此下去交通势必出现问题。

2》星和钻的信用等级问题：虽然在信用等级机制里面制定了“扣分”的机制。但因为评价质量已经被稀释，导致基本上不会出现星钻级别被扣的情况，全部都是在慢慢升级，只是上升的快慢不同。
到了一定高级别的时候卖/买家几乎可以高枕无忧了。
这就好像考了高分的同学可以不用好好学习，反正下次他最低还是那个分数。直接导致很多高等级的卖家开始出现交易质量问题，不思进取(据说这是淘宝现在的事实)。虽然无限长期来看这不是个严重的问题，但在实际情况来看这是个不小的问题。

3、信息机制是电子商务乃至整个社会的“游戏规则”，（我猜测）这个信用机制现在不仅是到买/卖家决定交易的参考值，也是淘宝推荐、搜索优先级等相关算法的参考值。如果真有问题，其长期危害比想象中要大很多。

4、产生这种问题的原因可能有很多。比如淘宝刚开始需要讨好卖家吸引买家、比如在运营、比如在市场、比如公关、比如….
产生这样的问题确实是产品策划者的责任，但这也是很多长线产品都会出现的问题。
眼前摆在淘宝面前最大的困难恐怕是：这些问题已成既定事实，无法从根上进行更改。一旦有大改动可能就会导致用户“造反”。（我不信淘宝有胆量把默认评价改成中评，更不信他们敢把降级机制降低）

5、那么能做的也只能优化或者稀释当前信用机制产生的“坏账”。
这就好像企业的管理，“对待能力不是很强或者工作不积极的老员工怎么办？”。 不能降他的级。此选择是把别人的整体级别提高，优一点的做法是给他空职让他做不是很重要的事。
我个人认为目前最合适的办法就是用“近期的信用”来稀释这个问题。（相当于给老员工空职）

6、 事实上对于淘宝来说星钻机制本身就是有问题的，因为这是一个“长期累计的信用”。
但，对于用户来说“近期的信用”比“长期累计的信用”更加重要，也更有参考价值。对于淘宝来说“近期的信用”也更能促进所有卖/买家的积极性。
设计者在设计信用机制逻辑的时候需要本着怀疑的态度。如果一人十年前是好人，那么他现在是好人的可能性有40%，如果一个人一天前是好人，那么他现在是好人的可能性有70%。

7、处理起来可以这样：
1》评价、星钻组成的“长期累计的信用”机制不变。
2》在界面上突出“近期的信用”，弱化“长期累计的信用”。
3》在搜索和推荐的机制算法上一步步的调整，让“近期的信用”在权重上慢慢提高。（这个算法要慢慢做偷偷做。免得有人造反）

8、不了解具体情况，很多事情只是猜测。仅为瞎扯。
不过通过这个问题，也许可以思考下长线产品的一些“游戏机制”问题。这些问题对于产品设计师来说永远是个很大的挑战，甚至是不可控的。

.

.

广告：
UCDChina正在寻找可以搞定语义分析和信息积累的程序高手 ，麻烦有类似资源的人帮忙推荐。