作者 Google 广告管理系统小组 李小轲

1. 使用系统管理网站旗下广告位置
2. 创建关联展示位置
3. 生成广告位置对应代码
4. 为广告主创建订单
5. 创建订单项 (1)

在订单创建完成之后，下面我们需要针对这个广告主的每一个系列创建单独的订单项。之前我的同事David曾经介绍过，对于网站所使用的广告投放系统，它的侧重于广告的排期和定位。这些操作都将在订单项这个界面完成，这里也是我们这个系统最核心的技术所在。下部我将向大家逐一介绍。

- 登录系统，点击“订单”，进入订单界面，选择并点击您需要添加广告系列的订单（如下图）



- 进入订单项界面，点击“新订单项”（如下图）



- 进入新订单项界面，下面我将按照功能模块分四个部分向大家介绍：

1) 基本：在“基本”模块中包括以下选项：

• 订单项名称：为订单项指定名称，如“奥林匹克系列”
• 投放优先级：当有多个广告订单在同一时间竞争同一广告位置时，系统会根据投放优先级来判断应该显示的广告。优先顺序从：独占，高级，标准，补余，内部，从高到低排列。独占订单项是唯一提供每日费用 (CPT) 定价方式的投放类型，所以如果您的广告订单希望以每日费用来进行结算的话，我们在这里可以选择“独占”，并在右边选择定价类型为“每日费用”。如果您的广告主希望以CPM或者CPC来进行结算的话，您可以选择高级或者标准。我们将在下一章中介绍“补余”和“内部”的使用。关于投放优先级的更多信息，您可以点击此处。
• 起始时间和结束时间：在此指定这个广告系列的时间跨度。所以针对奥林匹克系列，我们可以指定为2009年8月8日 – 2009年8月23日。针对秋冬系列，我们可以指定时间为2009年12月1日 – 2009年12月31日。
• 定价类型：在此指定广告订单的结算模式，如每日费用（CPT），千次展示费用（CPM），每点击费用（CPC）。注意只有“投放优先级”选择了“独占”后，才可以选择每日结算模式。
• 价格：对应每种结算模式，广告主承诺的结算价格
• 数量：此次广告订单的购买数量。要注意的事，这里指的是整个时间跨度的整体数量，而非每日数量。比如说，秋冬系列以CPM结算，跨越整个12月份的31天，希望购买1000万的展示量。在这里我们就需要输入1000万，而不是折算到每一天的展示量。

This is probably the feature I missed most from early days when I started to use InnoDB instead of MyISAM. Since that I figured out how to survive without it, but this is first question I hear from customers who migrated from MyISAM to InnoDB - can I just copy .ibd files from one server to another and answer “use mysqldump” is quite disappointed.
Jokes aside, I see real needs in this:
- when we need to restore only single table from backup (sometimes developers kill only single table, not whole database )
- to copy single table from production to QA environment. It may sound not so important, but I see needs in this quite often. QA boxes may have their own setup, not so powerful and with not enough space, but QA still needs to have some tables in fresh status.
- resharding databases, moving some shards from one server to another.

So long story short end - we made new mode for XtraBackup, now it can copy and prepare InnoDB/XtraDB tables, which later can be imported into XtraDB.

To be objective the process is still not so easy as with MyISAM tables when you just copy table.frm, table.MYD, table.MYI files, but so nice here - just can copy InnoDB tables in fully non-blocking mode, you do not need to shutdown neither source nor destination server (if destination server runs XtraDB).

In short, process looks like (ah, yes you also need innodb_file_table mode, which ones love, and ones hate)
1) do backup of specific tables on sources server with xtrabackup (server can run InnoDB or XtraDB, it does not matter)
2) do prepare of tables with –export options
3) create table on destination server with exactly the same CREATE TABLE statement as on source. (on destination you should have XtraDB with innodb_expand_import extension and with innodb_expand_import=1 setting)
4) run alter table IMPORTEDTABLE discard tablespace;
5) copy .exp and .ibd files to destination server
6) run alter table IMPORTEDTABLE import tablespace;
7) enjoy progress in error.log like:
InnoDB: import: extended import of test/img_out59
InnoDB: import: 3 indexes are detected.
InnoDB: Progress in %: 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 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 done.

Some more information available on
http://www.percona.com/docs/wiki/percona-xtradb:patch:innodb_expand_import

If you want some internals: I expected the biggest problem with implementation would be merging insert buffer or handling undo records, stored in ibdata file, but it was easy - you just need to shutdown instance with innodb_fast_shutdown=1 (done in xtrabackup –prepare –export call). The more complex problem was that InnoDB stores pointers on root leafs of secondary indexes also in system tablespace. And that’s why we maintain .exp files - to keep information from system tablespace that related to exported table.

Both xtrabackup –export mode and XtraDB innodb_expand_import available only in source code for now, you can get them from Launchpad projects https://launchpad.net/percona-xtradb and https://launchpad.net/percona-xtrabackup. Binary releases will be ready as soon as we have done with testing of this feature. You are welcome to try!

---

Entry posted by Vadim | No comment

Add to: | | | |

To start filtering out IPs, follow the steps below:

1. Collect IPs from anyone in your office (including yourself) that you don't want to track. If they don't know what their IP addresses are, an easy way to figure it out is to go to http://whatismyipaddress.com/.
2. Then, sign in to your Analytics account at http://www.google.com/analytics.
3. If you have more than one account, select the account that has the profile you want to apply the IP exclusion filter to.
4. Once you're on the Profile Overview page, click 'Edit' from underneath the Actions column.
5. From under the 'Filters Applied to Profile' section, select 'Add Filter.'
6. Select 'Add new Filter for Profile.'
   
   
   
7. Enter an easily identifiable Filter name (i.e. 'My IP address,' or 'CMO's IP address).
8. Select the filter type labeled 'Exclude all traffic from an IP address.' The IP address field will auto-populate with an example IP address. Enter the correct value. Remember to use regular expressions when entering any IP address. For example, if the IP address to filter is:
   
   176.168.1.1
   then the IP address value will be:
   176\.168\.1\.1
   
   
   
9. You may also enter a range of IP addresses. For example:
   Range: 176.168.1.1-25 and 10.0.0.1-14
   IP address value : ^176\.168\.1\.([1-9]|1[0-9]|2[0-5])$|^10\.0\.0\.([1-9]|1[0-4])$
10. Click ‘Save changes’ to finish.

Posted by Christina Park, Google Analytics Team

Recently Microsoft released a new search engine called Bing, which has begun sending traffic to websites with Google Analytics installed. Bing currently shows up as a referring site in the Traffic Sources reports. We are actively working on an update that will include Bing under the default list of search engines in Google Analytics.

If you absolutely can't wait for the upgrade, you can add the following line to your Google Analytics tracking code as a temporary solution. Note, this workaround is recommended for advanced users only. For more information on adding custom configurations to the Google Analytics tracking code, read this article.

pageTracker._addOrganic("bing", "q");

Stay tuned for updates.

Sebastian Tonkin, Google Analytics Team