Entry № 041-3 / V-3137 · 0:00 synced

World's Fastest Internet - 1.6 TERABITS per Second

Linus Tech Tips@LinusTechTips5.7M viewsJul 7, 20189:26
Source
YT
Views
5.7M
Subscribers
16.8M
Critic
?
Audience
?

0 up · 0 down · 0 ratings

Promos

Thanks to Com Hem for sponsoring this video! Learn more about their on-going collaboration with DreamHack at geni.us Get info about upcoming DreamHack events: geni.us Buy a good network switch On Amazon: geni.us On Newegg: geni.us Discuss on the forum: linustechtips.com Our Affiliates, Referral Programs, and Sponsors: linustechtips.com Linus Tech Tips merchandise at designbyhumans.com Linus Tech Tips posters at crowdmade.com Our Test Benches on Amazon: amazon.com Our production gear: geni.us Get LTX 2018 tickets at ltxexpo.com Twitter - twitter.com Facebook - @LinusTech Instagram - @linustech Twitch - twitch.tv Intro Screen Music Credit: Title: Laszlo - Supernova Video Link: youtube.com iTunes Download Link: itunes.apple.com Artist Link: soundcloud.com Outro Screen Music Credit: Approaching Nirvana - Sugar High youtube.com Sound effects provided by freesfx.co.uk

Start
AI OverviewDefault language

The video examines how DreamHack 2018 achieved a world record level of 1.6 terabits per second of internet bandwidth and what it takes to sustain that throughput in a large esports LAN party. It begins by setting the context that this speed is enough for every attendee in St John's Newfoundland to stream HD video concurrently and to download a major update like PUBG in a fraction of a second. The host travels to Sweden to tour the specialized gear, starting with a single yellow fiber cable that carries eight wavelengths of light, each capable of 200 gigabits per second. The explanation covers how passive and active signal manipulation separates the wavelengths, routing them into multiple fibers, and ultimately distributing the data to 6,000 plus participants through a network of 23 distribution switches and many access points. The video then explains how copper connections are used only for the final short distance, with 1 gigabit shared by most users unless a premium package upgrades to 10 gig uplinks. The system’s redundancy becomes a major focus, with dual routers at each end and multiple data center paths to ensure the network even remains functional if a main link is damaged. It shifts to automation and monitoring, describing an automated deployment system using DHCP, TFTP, and Python, supported by Prometheus for real-time health metrics and Grafana dashboards to visualize usage, hours played, and domain activity. Power demands are addressed with elaborate electrical infrastructure, including dedicated 10 kilovolt lines transformed to 400-volt rails, along with safety measures like RCDs to limit outages. Finally, the team demonstrates a custom broadcast solution that centralizes video and audio streams, enabling flexible input switching and in-game data overlays for moments like player health or bomb timers, all synchronized with stage lighting and pyrotechnics. The host closes by acknowledging the impressive integration of hardware, software, and event production that makes a mega-lan possible, and thanks Com Hem for sponsorship while inviting viewers to engage with DreamHack events.

Topics · technology · networking · esports · data_centers · energy_infrastructure

Questions answered

How is 1.6 terabits per second achieved over a single lan party network?
The speed comes from eight wavelengths of light inside a fiber cable, each capable of 200 gigabits per second. These wavelengths are separated by passive and active signal manipulation, then carried over a larger 16-fiber transmission path, converted to 100 gigabit signals for distribution, and finally delivered to users via a refined copper last mile with 1 gigabit or 10 gigabit uplinks for premium clients.
What redundancy measures keep the network up during a DreamHack event?
Two identical routers at each end provide failover, there are multiple paths from data centers to the floor, and a four 10 gigabit fiber backbone adds resilience. If a main link or router fails, the system switches to alternate routes so attendees maintain connectivity.
What roles do automation and monitoring play in managing the network at this scale?
Automated deployment using DHCP, TFTP, and Python configures each switch, while Prometheus monitors health and Grafana visualizes metrics like total hours played and network usage. This setup allows rapid detection and location tracking of failed devices, reducing manual intervention and downtime.