DaVinci Resolve & Cloud Computing: Using AWS For Video Editing

DaVinci Resolve & Cloud Computing: Using AWS For Video Editing

I’m breaking form my usual theme of working with data & databases. However this covers cloud computing and video editing which is a hobby of mine. As AWS offers Nvidia hardware, I want to leverage it to run DaVinci Resolve, my video editor of choice.

I travel from time to time (work and pleasure) and my old laptop isn’t up to running DaVinci Resolve anymore. It’s fine for everything else I want to do, but not video editing. This is both a numbers issue (Can AWS run DaVinci Resolve for a cost lower than or equivalent to buying a new laptop) and a technical feasibility issue (Can AWS run DaVinci Resolve successfully).

I’m not going to go into detail here on what AWS/EC2 (and the related technologies/products are) in this article to keep things concise.

Overview of Findings

Here are some of my key findings, which hopefully will inspire looking into the details below.

  • Yes, it’s feasible. AWS can run DaVinci Resolve. As long as it’s an EC2 instance with an Nvidia GPU.
  • Yes, it’s cost effective (for my case) to run DaVinci Resolve in the cloud.
  • You may have to think ahead/plan a workflow to accommodate using AWS (particularly allowing for uploading of source content).
  • You do need a good internet connection, with low latency for a comfortable workflow.
  • I have a dongle version of Resolve, I couldn’t get the dongle working via AWS. I don’t see any reason why a key based licence wouldn’t work (I’m happy to try one if someone will lend me a key).
  • Similarly, I couldn’t get my speed editor to work.

Cost Comparison

I have left some of the number crunching lower down on this page to show my working/thought process (for the dedicated). But I came to the conclusion that I’m estimating 5 – 10 days a year usage, over a lifetime of 5 years.

  • £3,000 for a new Nvidia Windows Laptop.
  • £1,000 for a new M1 Apple Macbook Air (cheapest option, being generous on price).

So my target is for less than £20 a day AWS costs (a £1,000 laptop, over 5 years, used 10 days a year). Which can be met using the NVIDIA Gaming PC – Windows Server 2019 (using the g4dn.2xlarge image, or the g4dn.4xlarge for shorter bursts)

Using AWS EC2 to run DaVinci Resolve

Technologies (products)

I’m going to focus on two key areas of the AWS catalogue (it’s massive, they have offerings and options than the biggest supermarket has breakfast cereals).

  1. EC2 (Elastic Compute Cloud): This is a virtual machine offering, allowing people to create as many virtual computers in the cloud as they like within minuets. Then thrown them away when they have finished with them, only paying for the time that the machine “exists”. Amazon EC2 – Amazon AWS
  2. S3 (Simple Storage Service): This is virtual disk storage, hosted by amazon. It’s commonly used for big data solutions, allowing users to create “buckets” to put data into. But it’s also a cost effective way of pushing small to extremely large amounts of data into the cloud and into the AWS ecosystem (ready for use by an EC2 instance. Simple Storage Service (Amazon S3) – Amazon AWS

EC2 is the main focus of this workflow, as I’m interested in leveraging the available AWS and NVIDIA virtual machines. DaVinci Resolve leverages the GPU, and having a nice powerful NVIDIA T4 GPU. Note, only DaVinci Resolve Studio (non free version) will fully utilise the GPU (although the free version gets a good portion of it).

AWS Configuration

To be able to run the G4dn Instances under EC2, I had to extend my quota: Amazon EC2 service quotas

To request an increase using the Amazon EC2 console

  • Open the Amazon EC2 console: https://console.aws.amazon.com/ec2/
  • From the navigation bar, select a Region.
  • From the navigation pane, choose Limits.
  • Select the resource in the list, and choose Request limit increase.
  • Complete the required fields on the limit increase form and choose Submit.
    • Configure Service Quota:
    • Running On-Demand G instances – I set this to 16 so I could use up to (but not accidentally more than) a 16 core machine.

AWS DaVinci Resolve Workflow

Data Upload (optional)

If you want to easily move data between S3 and your local machine/EC2 instance, follow these steps. If you want to just copy/paste within remote desktop (or use another method of transfer), then you can skip this:

  1. Configure Access Tokens: Before configuring your bucket, you will need these in the VM for Easy/command line access to S3 bucket. See tokens documentation: https://docs.aws.amazon.com/general/latest/gr/aws-sec-cred-types.html#access-keys-and-secret-access-keys
  2. Create Bucket: Follow S3 instructions
  3. Upload Data: Follow S3 instructions

Create VM

You will want to use one of these, it’s a preconfigured server image for the Nvidia hardware, you can select the number of vCPU cores and amount of RAM:

Select VM type, I recommend filtering for g4dn and selecting either the 8 or 16 core machine (more money = more power).

You will need to get the admin password (wait 4 mins after creation of VM).

Connect To VM

Once you have the hostname, username and password, you can connect using Windows Remote Desktop (search for mstsc in the start menu if you can’t find it)

Note: the AWS console does allow you to download a configuration file with the username and hostname which you can use to launch the application rather than copy and paste into the application.

If you are reconnecting:

  1. Check the firewall rules, by default the setup uses the IP address from which you first connected. So, if you connect from a new outside location (or your home IP changes) the firewall will block you.
  2. Check the Public DNS. If you stop, then restart the EC2 instance, the IP address (and Public DNS associated with it) will change.

Configure VM & Install Software

DaVinci Resolve fails to install (thanks to the workings of default PostgreSQL installation). As this isn’t a production world facing server, we can cut back the security. Use the policy editor in the VM to set:

  1. Local Security Policy (Editor)
    1. Account Policies → Password Policy: Password must meet complexity requirements = Disabled
  2. Local Group Policy Editor
    1. Computer Configuration → Administrative Templates → Windows Components → Internet Explorer → Internet Control Panel → Advanced Page → Turn On Enhanced Protected Mode = Disabled

Other Configuration steps:

  1. Format/create hard disk
    1. Note: If you “stop” the VM to reconnect again, the disk D will be gone
  2. Enable “Windows Audio” service (set to start on startup)

Software installation:

  1. AWS CLI (for S3)
    1. https://docs.aws.amazon.com/cli/latest/userguide/install-cliv2-windows.html
    2. https://docs.aws.amazon.com/general/latest/gr/aws-sec-cred-types.html
  2. I use ninite.com to install extra applications quickly including:
    1. Notepad++
    2. peazip
    3. VLC
  3. Download resolve: https://www.blackmagicdesign.com/products/davinciresolve/

Optional Session Manager:

Data Import

Note: If you “stop” the VM to reconnect again, the disk D will be gone!

Copy via:

  • RDP
  • S3 – This is where you log into S3, or you use the tokens previously configured.

Get Creating

You are now free to work as normal, using a powerful cloud based computer for the processing. Just remember:

  1. Power-off the VM when you are done, you pay for the time it’s running.
  2. Backup all your work to S3 or local storage as the D drive on the VM gets destroyed when you power-off the VM.

Crunching the Numbers

Feel free to skip this bit, it’s the full breakdown of my numbers and calculations, I’ll summarise the findings at the end.

Laptop Costs

Being a data guy, I built the below table to see how much cost/value I’d get (right now, video editing is my only requirement for new hardware). If I spent £1,000.00 on a laptop and I used it for video editing 10 days in a year over a lifespan of 5 years before replacing, it would be equivalent to £20 a day. Worst case scenario is I spent £3,000.00 on a laptop and replace it after 3 years only doing video editing on average of 5 days a year £200 a day!

CostYearsCost per yearDays per yearCost per day
£1,000.005£200.0010£20.00
£2,000.005£400.0010£40.00
£3,000.005£600.0010£60.00
£1,000.003£333.3310£33.33
£2,000.003£666.6710£66.67
£3,000.003£1,000.0010£100.00
£1,000.005£200.005£40.00
£2,000.005£400.005£80.00
£3,000.005£600.005£120.00
£1,000.003£333.335£66.67
£2,000.003£666.675£133.33
£3,000.003£1,000.005£200.00

The numbers contain a lot of “if”s and of course at any time there can be other reasons I want to upgrade my laptop. But for now, if I can find an alternative which is cheaper than £20 a day I win all around. So, now I have my benchmark to aim for I can look into AWS costs (which vary over time and region).

EC2 Costs

Here are some sample costs for EC2 (the actual virtual computer). Note, when not running the OS hard disk is charged as S3 data storage, and the EBS disk is destroyed (no costs). So, the non running cost of this is under £5 a month.

Modelg4dn.xlargeg4dn.2xlargeg4dn.4xlarge
Cores4816
Memory (GB)163264
Hour Cost (USD)0.7991.2472.144
8 Hour Day (USD)6.3929.97617.152
10 Hour Day (USD)7.9912.4721.44
12 Hour Day (USD)9.58814.96425.728

In reality, few cases will require the g4dn.4xlarge, and most will be perfectly happy on the g4dn.2xlarge instances. Which is much more fitting (even for a 12 hour day) with the target.

S3 Storage Costs

This can clearly get pricy over time, but for short sprints of work the cost is negligible. It’s also billed to smaller units of time even if priced monthly. So, if you only use the storage for a day or two, it’s only a few cents. All costs below in USD, based off pricing for London.

S3 StandardS3 Standard
Infrequent Access
EBS SSD
(gp3) – Storage
GB Cost (per month) 0.0240.01310.0928
GBs200200200
Monthly Cost4.802.6218.56
Yearly Cost57.631.44222.72

For 200 gigs of data, that cost will mount up over the year, but month by month it’s pretty negligible and the standard storage of £57.60 is still a tiny proportion of the laptop costs above.

Cost Summary

As the tables above show, there is a lot of room for variance and options. But for 10 days a year average usage of video editing while away from home, using cloud computing is a very viable alternative to buying a new laptop for the foreseeable future.

A poor case example is a long ongoing project (all year) with 200 gigs storage (£57.60), the virtual machine always available (12 months @ £5 = £60) and 10 days working a 10 hour day on a 4xlarge machine (10 * £21.44 = £214.4) Gives £332 for the year. Equivalent to a £1,000.00 laptop used for 5 days a year, cheaper than a £2,000.00 used for 10 days a year.

In reality, most projects will require an instance for a long weekend, and the storage can be thrown away after returning home and incorporating it. There are other costs to be aware of, but they are comparatively negligible.

More details

For more information/some useful links:

2 Replies to “DaVinci Resolve & Cloud Computing: Using AWS For Video Editing”

    1. Sorry for the slow response (you got caught in the spam filter), I haven’t tested it with panels but I couldn’t get it to work with the USB dongle so I think you might be out of luck for this particular method. Instead, I’d recommend looking into the official AWS image from Blackmagic Design (created after I started work on this post) https://aws.amazon.com/marketplace/pp/prodview-zzy5tef4cq6sg

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.