想做的事情,應該做的事情,不要管別人如何說,去做就對了,
把事情的結果這個頭痛糾結的問題就交給阿爸父吧。
Apache is the most popular web server in use today because it is free, runs everywhere, performs well, and can be configured to handle most needs.
http://httpd.apache.org/">Apache
Linux is a very popular OS in data centers because it is free, runs on a lot of hardware, has tons of available software, highly performing, easily virtualizable, and flexible. All good attributes when you are starting a web site and hoping to grow with demand.
Some popular versions of Linux used in data centers are: CentOS, Red Hat, and Ubuntu.
http://www.linux.org/">Linux
http://www.mysql.com/">MySQL
http://highscalability.com/lighttpd">lighttpd
for video instead of Apache
while (true)
{
identify_and_fix_bottlenecks();
drink();
sleep();
notice_new_bottleneck();
}
This loop runs many times a day.
http://linux.die.net/man/4/epoll">epoll
to wait on multiple fds.CDN is a system of computers networked together across the Internet that cooperate transparently to deliver content (especially large media content) to end users. The first web content based CDN's were Sandpiper and Skycache followed by Akamai and Digital Island. The first video based CDN was iBEAM Broadcasting.
CDN nodes are deployed in multiple locations, often over multiple backbones. These nodes cooperate with each other to satisfy requests for content by end users, transparently moving content behind the scenes to optimize the delivery process. Optimization can take the form of reducing bandwidth costs, improving end-user performance, or both.
The number of nodes and servers making up a CDN varies, depending on the architecture, some reaching thousands of nodes with tens of thousands of servers.
http://en.wikipedia.org/wiki/Content_Delivery_Network">CDN
(content delivery network):http://en.wikipedia.org/wiki/Colocation">colo
sites.http://en.wikipedia.org/wiki/RAID
">RAID
http://labs.google.com/papers/bigtable.html">BigTable
, a distributed data store:http://www.possibility.com/epowiki/Wiki.jsp?page=DatacenterSystemChoiceAnalysis">Data Center
StrategySome advantages are:
* faster backup
* faster recovery
* data can fit into memory
* data is easier to manage
* provided more write bandwidth because you aren't writing to a single master. In a single master architecture write bandwidth is throttled.
This technique is used by many large websites, including eBay, Yahoo, LiveJournal, and Flickr.">Shard
. Sharding helps to isolate and constrain storage, CPU, memory, and IO. It's not just about getting more writes performance.