Ever wonder how Google stays so reliable and quick? Did you know that they use a stripped down version of Red Hat? Did you know they built their own file system? This article explains it all.
“We take our files and chunk them up, then you randomly distribute the chunks across different machines, making sure each chunk has at least two copies that are not physically adjacent — not on same power strip or same switch,” Hoelzle said. “We try to make sure that even if one copy goes away, another copy is still here.” Chunks typically are 64 megabytes and are replicated three times.
All this replication makes it easier to make changes, Hoelzle said. Google simply takes one replica at a time offline, updates it, then plugs the machines back in.
This is a very well written and interesting read on how Google serves up information fast and accurately. You’ll find it both techinical yeat written in a way you should be able to understand pretty easily.
Tags: Google, Uncategorized
