The internet contains roughly 64 zettabytes of data as of 2024, though this number grows constantly and varies depending on what you count

A zettabyte is one sextillion bytes — that's a 1 followed by 21 zeros. To put it in perspective: one zettabyte equals about 1 trillion gigabytes, or roughly 250 million DVDs worth of storage. The 64-zettabyte figure comes from industry analysts like IDC and Statista, who track data creation across servers, devices, and cloud storage worldwide.

But this number is slippery. It counts data that exists somewhere — stored on servers, in data centers, on your phone, in email inboxes — not data you can actually access. Much of it is duplicated (your email provider keeps backups). Some is encrypted or locked behind paywalls. Some is deleted but still technically recoverable. The "size of the internet" depends entirely on what you're measuring.

The number also grows faster than most people realize. Data creation accelerated sharply after 2015, when smartphones became ubiquitous and cloud storage became cheap. Video streaming alone — Netflix, YouTube, TikTok — now accounts for roughly 60 to 80 percent of all internet traffic, depending on the region and time of day.

Key Takeaways

  • The internet contains roughly 64 zettabytes of data, but this includes duplicates, backups, and data you cannot access.
  • Video streaming makes up the majority of internet traffic and data storage, far more than text, images, or social media.
  • Data creation is accelerating — the amount of data created each year now exceeds the total amount created in all previous years combined.
  • Most of the internet is not the public web; it includes private corporate networks, cloud backups, email servers, and data centers you never see.
  • Search engines index only a tiny fraction of the internet, so most data is invisible to Google and other search tools.

What counts as "internet data" and what does not

When analysts measure the size of the internet, they typically count all digital data stored on networked computers worldwide — not just the public web. This includes your email inbox, your cloud backups, Netflix's video library, your bank's transaction records, and corporate databases that never touch the public internet.

The public web — the part you can reach with a search engine — is a fraction of this. Google indexes roughly 5.6 billion web pages, which sounds enormous until you realize it represents less than 0.004 percent of all internet data. The rest is the deep web: password-protected email, medical records, financial accounts, academic databases, and private company networks. None of this is hidden or illegal; it's simply not indexed by search engines because it requires authentication.

The measurement also includes redundancy. If you upload a photo to Google Photos, Google stores multiple copies across different data centers for reliability. If you send an email, it exists on your device, your email provider's server, and your recipient's device. These are counted as separate data points, even though they represent one piece of information.

Where all this data actually lives

Most internet data lives in data centers — massive buildings filled with servers, cooling systems, and backup power supplies. The largest operators are Amazon Web Services (AWS), Microsoft Azure, Google Cloud, and Alibaba Cloud. These companies own or lease thousands of data centers worldwide.

Your personal data is scattered across multiple locations. When you use Gmail, Google stores your emails on servers in multiple countries for redundancy. When you stream Netflix, the video is served from a server geographically close to you to reduce lag. When you use iCloud or OneDrive, your files are backed up to data centers you'll never see.

A smaller portion lives on personal devices — your phone, laptop, external hard drives. But even this is increasingly synced to the cloud, so it exists in both places simultaneously. The shift toward cloud storage means less data lives solely on devices you own.

How fast the internet is growing

Internet data doubles roughly every two to three years, though the rate varies by source. IDC estimates that 120 zettabytes of data will be created in 2024 alone — meaning more data will be created this year than existed on the entire internet in 2020.

This acceleration is driven by a few specific things. Video streaming generates enormous amounts of data: a single hour of 4K video is roughly 25 gigabytes. Social media platforms like TikTok and Instagram generate billions of photos and videos daily. Internet-connected devices — security cameras, smart home systems, fitness trackers, industrial sensors — constantly transmit data. A single self-driving car generates roughly 1 terabyte of data per day.

Not all of this growth is new information. Much of it is redundancy, compression, and storage of the same content in multiple formats. A video might be stored in 4K, 1080p, and 720p simultaneously to serve different devices. Backups and archives account for a significant portion of total data.

Why most of the internet is invisible to search engines

Google, Bing, and other search engines use automated programs called crawlers to discover and index web pages. But crawlers cannot access password-protected content, and they respect instructions from websites telling them not to index certain pages. This means the vast majority of internet data is never catalogued by search engines.

Academic databases, medical records, financial accounts, and corporate intranets are all part of the internet — they use the same underlying technology — but they're not searchable. This is intentional. You don't want your bank account visible to Google. Neither do hospitals, law firms, or companies protecting proprietary information.

The indexed web — what you can find through search — is estimated at 5 to 10 billion pages depending on the source. The deep web is estimated to be 400 to 500 times larger. This doesn't mean the deep web is dangerous or full of illegal content; it's simply the parts of the internet that require login credentials or are intentionally hidden from search engines.

How data size affects your privacy and security

The sheer volume of internet data creates both privacy risks and protections. On one hand, more data means more opportunities for breaches. A data center storing millions of records is a larger target than a small server. On the other hand, the scale of the internet makes it harder for any single actor to monitor all of it.

Your personal data is a tiny fraction of the total — probably a few gigabytes at most across all your accounts and devices. But that data is valuable because it's concentrated in a few places: your email provider, your cloud storage, your social media accounts. A breach at one of these companies can expose millions of people's data simultaneously.

Understanding the scale of the internet also helps explain why privacy policies matter. When a company collects data from billions of users, even small percentages represent enormous amounts of information. If a platform has 2 billion users and collects just 1 megabyte of data per user per month, that's 2 exabytes of data annually — more than the entire Library of Congress.

The difference between data storage and data transfer

When people talk about "the size of the internet," they usually mean stored data — how much information exists on servers and devices. But the internet also moves data constantly. Bandwidth measures how much data flows through the internet per second.

Global internet traffic is roughly 4.8 zettabytes per year as of 2023, according to Cisco. This is different from stored data. Traffic is the flow; storage is the pool. You might have 100 gigabytes stored in cloud backup, but transfer only 1 gigabyte per month to and from it.

Video streaming dominates traffic because video files are large and people watch continuously. A single Netflix subscriber might transfer 10 to 50 gigabytes per month depending on viewing habits and video quality. Multiply that by 250 million subscribers worldwide, and the bandwidth required is enormous.

Frequently Asked Questions

Is the "dark web" part of the 64 zettabytes?

The dark web — sites accessed through Tor or similar anonymity networks — is technically part of the internet, but it represents a tiny fraction of total data. Most estimates suggest it's less than 0.01 percent of all internet data. The vast majority of the internet that's invisible to search engines is simply password-protected, not intentionally hidden.

How much of the internet is actually useful information versus duplicates and backups?

This varies widely and is difficult to measure precisely. Industry estimates suggest 20 to 40 percent of stored data is redundant backups and duplicates. The rest includes active files, databases, cached content, and archives. Much of it is useful, but the same information often exists in multiple places.

Does deleting a file from my device remove it from the internet?

Not necessarily. If you uploaded it to cloud storage, it still exists there unless you delete it from the cloud service too. If you emailed it, copies exist on email servers. If it was posted publicly, it may have been cached or archived by services like the Wayback Machine. Deletion is local to your device unless you actively remove it from all places it was stored.

Why does the internet feel slower if there's so much data available?

Speed depends on bandwidth and distance, not total data size. You might have access to petabytes of data, but if your internet connection is 50 megabits per second, you can only download 6.25 megabytes per second. Data centers also prioritize traffic, so popular content loads faster than obscure content.

Is the 64-zettabyte number accurate, or is it a guess?

It's an informed estimate based on data from major cloud providers, internet service providers, and device manufacturers. Different research firms produce slightly different numbers — some say 60 zettabytes, others say 70 — because they measure different things and use different methodologies. The exact number matters less than understanding that it's enormous and growing rapidly.