Discovering Data Centers
Season 1, Episode 3
Season 1, Episode 3
“What are the core principles behind Google data centers?”
In this episode, we demonstrate how performance, availability, and security are implemented in Google’s data center infrastructure.
Watch
Link to Youtube Video (visible only when JS is disabled)
What are the core principles behind Google data centers?
[00:07] at google data centers we have a relentless focus on performance our data centers are more than just buildings with
[00:14] a collection of machines wired together they house clusters of hundreds of thousands of servers in locations across the globe
[00:21] that need to act as a unified network so what does it take to operate and build them efficiently at
[00:27] massive scale it all comes down to four core principles performance availability security and sustainability to optimize for performance we
[00:39] don't rely on off-the-shelf components we customize the design and build of almost every part of the stack for ultra
[00:45] high performance this includes electrical substations servers racks and even how we operate cooling plants you might think that means
[00:55] we're using high-end computers but at google we build from the bottom up commodity hardware underpins our custom tech stack
[01:04] that runs hundreds of thousands of jobs across these machines to give you distributed performance and scale but this doesn't
[01:11] mean we're using the desktop computer you're used to at home although your computer has a lot of technology that's
[01:16] optimized for how you use it our data center servers use only the components needed for lean high performance when
[01:24] multiplied by the thousands we provide you a sweet spot between performance and cost google cloud servers are built using
[01:31] the google titan chip from the time of manufacturing titan helps secure every server through its entire lifecycle it uses
[01:39] root of trust technology which cryptographically ensures that the chip hasn't been tampered with and significantly reduces the chance of
[01:46] vulnerabilities but what happens if a machine fails billions of users depend on our services being up and running 24/7
[01:55] so google uses a fault tolerant design that's maintainable from concept to operations at the infrastructure level this boils
[02:03] down to constant monitoring of every hardware component electrical and mechanical system for configuration activity environmental and error data we
[02:15] use machine learning and machine failure diagnostic tools to suggest corrective actions and because we're constantly upgrading and reconfiguring our
[02:24] infrastructure our hardware operations teams do deployments maintenance upgrades and repairs of all hardware 24/7. this all comes with
[02:34] the key responsibility of protecting and securing our data centers for every user and google cloud customer more on that
[02:41] next time on discovering data
[00:07] at google data centers we have a relentless focus on performance our data centers are more than just buildings with
[00:14] a collection of machines wired together they house clusters of hundreds of thousands of servers in locations across the globe
[00:21] that need to act as a unified network so what does it take to operate and build them efficiently at
[00:27] massive scale it all comes down to four core principles performance availability security and sustainability to optimize for performance we
[00:39] don't rely on off-the-shelf components we customize the design and build of almost every part of the stack for ultra
[00:45] high performance this includes electrical substations servers racks and even how we operate cooling plants you might think that means
[00:55] we're using high-end computers but at google we build from the bottom up commodity hardware underpins our custom tech stack
[01:04] that runs hundreds of thousands of jobs across these machines to give you distributed performance and scale but this doesn't
[01:11] mean we're using the desktop computer you're used to at home although your computer has a lot of technology that's
[01:16] optimized for how you use it our data center servers use only the components needed for lean high performance when
[01:24] multiplied by the thousands we provide you a sweet spot between performance and cost google cloud servers are built using
[01:31] the google titan chip from the time of manufacturing titan helps secure every server through its entire lifecycle it uses
[01:39] root of trust technology which cryptographically ensures that the chip hasn't been tampered with and significantly reduces the chance of
[01:46] vulnerabilities but what happens if a machine fails billions of users depend on our services being up and running 24/7
[01:55] so google uses a fault tolerant design that's maintainable from concept to operations at the infrastructure level this boils
[02:03] down to constant monitoring of every hardware component electrical and mechanical system for configuration activity environmental and error data we
[02:15] use machine learning and machine failure diagnostic tools to suggest corrective actions and because we're constantly upgrading and reconfiguring our
[02:24] infrastructure our hardware operations teams do deployments maintenance upgrades and repairs of all hardware 24/7. this all comes with
[02:34] the key responsibility of protecting and securing our data centers for every user and google cloud customer more on that
[02:41] next time on discovering data