How Much Space Is Uploaded To Youtube Every Minute

Clarifications:

i. Includes advertising videos? Lets say No

2. How much celebrated information is chosen – you decide

3. Cache / CDN included – Aye

Approach:

1. Bottom-upward estimation of content uploaded past creators based on Google announcement benchmarks

two. Boilerplate out the content in the last 10 years based on best known stats

3. Look at data replication strategies to see low latency/HA and assign content duplication buffers

Assumptions:

i. 500 Hours video uploaded every minute to YT à 500 x 60 x 24 = 720K = 0.72 Million hours of video upload per day (Ggl I/O announcement 2019)

2. 40 ZB of information was created globally in 2019 (statista written report)

3. Global digital data creation grew at xx CAGR from 2011 and 2021 (statista report)

Working:

one. Determining the size (per unit of fourth dimension) of a unmarried uploaded video

a. Resolution = 480p @ 9:16 aspect ratio = (480 ten 480 x (xvi/9) = 400K pixels

b. Color-depth = xvi bits / pixel

c. Size of i frame = 400K x xvi = 6.4 Mbits

d. Frame-rate = thirty frames/second

e. Per second raw video = 6.4 x 30 ~ 200 Mbps

f. Compression with standard video encoding methods = 1:160

k. Total bit-stream rate = 200 Mbps / 160 = 1250 Kbps (Its ok to assume this effigy direct during interview every bit these are kind of standard numbers)

h. This is aforementioned as 1250 10 one thousand x 3600 / 8 = 0.5 GB / 60 minutes

2. In similar fashion:

a. 720p = 1.5 GB/Hr

b. 1080p (Full Hard disk drive) = 2.5GB/Hour

c. 4k = 12GB / Hr

iii. As of stat published past Google in 2019, 500 Hours of video is uploaded every minute

a. This means 500 x 60 = 30K Hours of video uploaded / Hour

b. Which is aforementioned every bit 720K Hours of videos uploaded every 24-hour interval

four. Lets brand some assumption for the separate of this content across resolutions

Resolution	% of content (based on observation)	Size / hour of video (come across ane and two above)	Apportioned Uploaded data / twenty-four hour period
4K	1%	12 GB / Hr	720 K * 0.01 * 12 = 86 TB
Full-HD (1080p)	nineteen%	2.5 GB/Hour	720 1000 * 0.19 * 2.5 = 340 TB
720p (Standard)	50%	1.5 GB / Hr	720 Chiliad * 0.5 * i.5 = 540 TB
480p	30%	0.5 GB / Hr	720 K * 0.iii * 0.5 = 108 TB
Total			1074 TB ~ 1 Atomic number 82

1. ane PB of data is uploaded per mean solar day in 2019

2. This means ~350 Lead of information in 2019

3. Quick Sanity Bank check at this point

a. 40 ZB of Global digital information was created in 2019

b. And then total YT content created in 2019 is 40000:0.three ~ 120K:1 fraction of total digital content created globally which seems to exist in right order of magnitude looking at various factors like YT creator % etc

4. Between 2012 and 2021, global digital information grew at 20% CAGR

five. Lets presume the same pattern for YT information that's uploaded. Further let's assume that YT maintains information for 10 years. With that info we tin deduct the information for the 10-year flow of 2012 – 2021 with the formula:

End Value = Begin Value (ane + CAGR/100)^n

(here first derive begin value for 2012 past setting finish value for 2019 as 350 and n = seven then use this value to derive all other year number.

Year	Data Uploaded (PB)
2012	98
2013	117
2014	141
2015	169
2016	203
2017	243
2018	292
2019	350
2020	420
2021	504
Total	~2500

This ways for concluding 10 years ii.5 EB of data is uploaded to YT which is stored.

1. At present permit united states endeavor to look deeper into how YT will shop this information as it will need to ensure distributed and many a times redundant storage to ensure low latency and high speed streaming.

2. For that lets look at a high level abstraction of a possible layout of YT content storage.

Of import is to notation that the video uploaded by creator community is stored in a blob store like the GCP Cloud Storage. But it is so processed – separate into chunks for streaming, pre-candy for multiple device, multiple resolution, multiple bandwidth options, compressed and then partially / on-demand replicated to a globally distributed CDN so that its optimized for viewing across a huge a global consumer community of thirty K DAUs

one. We presume that this compression + global replication (not total equally there will be enshroud misses) will lead to at to the lowest degree 3X more data than whats originally uploaded

2. This ways full storage = Blob Storage + Popular distributed CDN cache (Ggl + 3rd political party)

3. This will sum upto 4 ten 2500 Pb = 10 EB data.

four. This will exist the full storage requirement of YT