t**r
发帖数: 3428 | 1 udemy 的casandra course有人上过么？好么？ |
|
h*********e
发帖数: 247 | 2 Location: LA
H1b/Green Card support
1. Critical Thinking Software Engineers (All Levels)
We are not a Java shop. We are not a Ruby shop, We are not a PHP, Perl or
Python shop either. We deploy the technology that best addresses the needs
of our businesses. Casandra, Hadoop, noSQL, MongoDB and Oracle are also all
present in our stacks.
As such we are looking for true software engineers that are good thinkers
and above average problem solvers. We are looking for engineers that have a
real passion ... 阅读全帖 |
|
h*********e
发帖数: 247 | 3 【 以下文字转载自 JobHunting 讨论区 】
发信人: happychoice (快乐公主), 信区: JobHunting
标 题: IT公司招人，找工的兄弟姐妹看过来啦！
发信站: BBS 未名空间站 (Sun Mar 11 20:52:56 2012, 美东)
Location: LA
H1b/Green Card support
1. Critical Thinking Software Engineers (All Levels)
We are not a Java shop. We are not a Ruby shop, We are not a PHP, Perl or
Python shop either. We deploy the technology that best addresses the needs
of our businesses. Casandra, Hadoop, noSQL, MongoDB and Oracle are also all
present in our stacks.
As such we are looking f... 阅读全帖 |
|
I*******g
发帖数: 7600 | 4 【 以下文字转载自 JobHunting 讨论区 】
发信人: IFloating (Floating Freely), 信区: JobHunting
标 题: 现在start up 铺天盖地
发信站: BBS 未名空间站 (Thu Sep 3 22:12:53 2015, 美东)
说到底， 就是因为太多的open source拿来就用， 没有成本，
都是一窝蜂， 最后98.9%都死在沙滩上。
比如，mongoDB, casandra, Zookeeper, Hadoop, Hive, Apache Kafka, Storm,
Couchbase and Vertica |
|
c*****g
发帖数: 21627 | 5 哈哈哈，我刚给阿三在校生面过，
尼玛，
10个阿三里，至少7个阿三
简历上名词乱堆，什么hadoop hive casandra，一问三不知当然要充数
所谓的经验就是做做测试，run run 别人写的script，最后统计性能
和编程一点关系都没有
当然，真正有经验的阿三也有 |
|
I*******g
发帖数: 7600 | 6 说到底， 就是因为太多的open source拿来就用， 没有成本，
都是一窝蜂， 最后98.9%都死在沙滩上。
比如，mongoDB, casandra, Zookeeper, Hadoop, Hive, Apache Kafka, Storm,
Couchbase and Vertica |
|
c*******a
发帖数: 1879 | 7 大家都是怎么用的呢？
光存储 不能快速搜索用也没有用啊
FB 开源它 就是一个装逼犯 |
|
|
|
|
d********g
发帖数: 42 | 11 You miss the point of using cassandra, it is not meant to be fast in
searching in the first place. BigTable/Hbase is the same. |
|
u***n
发帖数: 21026 | 12 那要search 功能怎么办？data migrate去RDS？ |
|
|
u***n
发帖数: 21026 | 14 那data放哪里啊，es也不是要query数据库的，我一直没有搞明白elastic search，因
为我也不怎么用 |
|
|
|
m****s
发帖数: 18160 | 17 【 以下文字转载自 Database 讨论区 】
发信人: jelty (TT is refreshing), 信区: Database
标 题: DBA opening in Bay area and Boston
发信站: BBS 未名空间站 (Thu Apr 10 01:17:40 2014, 美东)
you can send me your resume to j*******[email protected] or apply directly through
our company website
About Lattice
Lattice is an early innovator in one of the hottest areas of the marketing
technology market: predictive applications. We bring every relevant buying
signal into the industry’s most comprehensive cloud-based suite of business
appli... 阅读全帖 |
|
M*****r
发帖数: 1536 | 18 http://www.unixmen.com/news-today/875-digg-says-yes-to-nosql-and-bye-to-mysql
12 March 2010
Posted in News - News
After twitter, now is Digg who decided to replace MySQL and most of their
infrastructure components and move away from LAMP to another architecture
called NoSQL that is based in Casandra, an opensource project that develops
a highly scalable second-generation distributed database. Cassandra was open
sourced by Facebook in 2008 and is licencied under Apache Licenses..
The reason of |
|
|
g*****g
发帖数: 34805 | 20 Biggest PAAS player, Salesforce, uses Java as their main language.
Biggest SAAS deployment in the cloud, Netflix, uses Java as their main
language.
I think cloud technologies also drive the NoSQL DBs, with Casandra taking
the lead, a pure Java DB. Hadoop is the most popular framework for big data
processing, again, written in Java.
Cloud computing makes Java an easy choice. Linux dominates there. .Net doesn
't run on Linux. The most well known tools are already written in Java.
Enough said. |
|
g*****g
发帖数: 34805 | 21 当然是100%保持，如果你不知道cassandra是怎么工作，自己去读一读吧。
写一个结点能到1000，写100个就可以接近10万。
你丫就这点货，还有脸吹一个DB throughput秒杀Casandra。完全不懂得scale out怎么
回事。 |
|
T********i
发帖数: 2416 | 22 goodbug和zhaoce都是输不起的。而且辩论的品格非常不好。本来不打算回应就算了。
之所以回应就是因为害怕谬种流传。很多不明真相的网友把谬误当作真理。当然也包括
这两位。
goodbug号称用cassandra方案解决我提出的春运售票问题。在我提出我的方案
performance的throughput比他的高1-2个数量级的情况下。他认为他的方案可靠性比我
的好。理由是能够同步写log到磁盘。在我提出用磁盘（磁盘是机械式的，SSD不是磁盘
，是基于半导体的）同步写每秒不到千的情况下，他给我一个链接和一张图。号称【
quote】写一个结点能到1000，写100个就可以接近10万。【quote】
我只能这种人的基本功连做一个普通程序员都不见得够格。如何能有人追捧他？看来公
共论坛水的嗓门大貌似水有理。
简单说一下：
1. 操作系统下普通的文件写操作一般是写到缓冲区，然后由OS schedule磁盘的写操作
。要强制flush缓冲区需要特殊API fsync。这个过程很费时，而且是blocking的。因为
普通磁盘的寻道时间都在5ms以上。IOPS(IO per second)不会很高... 阅读全帖 |
|
g*****g
发帖数: 34805 | 23 我也觉得很奇怪，Cassandra值5亿，这比Casandra快1-2数量级的东西怎么也值10亿。
但魏公公吹完牛皮，下面就没有了。 |
|
|
|
