Infrastructure & Data – Transcoding at scale · cloud-based video encoding with GPU servers

Video encoding: transcoding at the pace of uploads

FFmpeg pipelines with GPU acceleration: convert video uploads into every format and resolution – scaled by queue, not by wishful thinking. In times of high-speed internet and ever more powerful devices, the demands on video productions are rising rapidly: users expect razor-sharp images, smooth streaming and maximum compatibility on all platforms – for smart workflows with more iterations and faster rendering from previsualisation to the final frame.

ccloud³ Console · encoding farm
eu-de · Hallstadt Data Centre
Queue
38 jobs
Workers
2 → 8 GPU
Throughput
14× real time
JOB upload_4k.mov · ABR set
● Encoding
GPU NVENC · RTX A4000
● Busy
S3 Renditions · object storage
● Stored
K8S AutoScaler · worker pool
● Scaled
Queue processedworkers · scaled to 0
  • Maximum video quality without compromise – razor-sharp images, smooth streaming and maximum compatibility on all platforms.
  • Access to the latest hardware – Nvidia RTX A4000, RTX 6000 Ada and A100, always up to date for maximum performance and best image quality.
  • Efficient distribution of computing resources – additional instances are available automatically under higher data load, especially for large-scale streaming projects.
  • Significantly reduced total costs – rent infrastructure only for the period you really need it instead of buying and maintaining servers.
Why centron

The on-demand encoding farm

Encoding is predictably computationally intensive – ideal for infrastructure that scales with the queue.

NVENC acceleration

GPU encoding on RTX graphics cards converts streams many times faster than CPU-only encoding – from €92.59 per month (billed by the hour).

Worker by Queue

Kubernetes starts encoding workers when jobs are queued and stops them afterwards – no idle costs.

In and Out via S3

Source files and finished renditions are transferred via S3 – with free traffic for delivery as well.

Jobs that are easy to follow

Job queues and monitoring display throughput, errors and costs for each encoding profile.

Use cases

The solution: cloud-based video encoding

Bandwidth limits, network congestion, device fragmentation, storage costs and choosing the right codec are the challenges of modern video producers – cloud-based video encoding solves them with hardware on demand.

Why video encoding is so important

Professional video encoding converts your raw video data into highly compressed, compatible formats. This enables shorter loading times, smaller data volumes and optimal display on different devices – whereas your own encoding infrastructure demands expensive hardware, ongoing maintenance and specialist knowledge.

Access to the latest hardware

We always keep our hardware up to date. That guarantees maximum performance and the best image quality – GPU servers with Nvidia RTX A4000, RTX 6000 Ada and A100 as Cloud GPU, billed by the hour.

Efficient distribution of computing resources

Especially for large-scale streaming projects, the cloud scores with scalable performance: additional instances become available automatically under higher data load – encoding workers in Kubernetes follow the queue, results go to video hosting.

Significantly reduced total costs

Running your own servers is expensive – in purchase as well as in maintenance and updating. With cloud services you simply rent the infrastructure you need for the period you really need it – with predictable budgeting.

From upload to ABR set

Your encoding pipeline, automated

An upload triggers the job; FFmpeg generates the complete ABR set (1080p to 240p plus thumbnails), and the results are stored in S3 with versioning – ready for streaming. The AutoScaler handles peaks following events by deploying additional GPU workers.

GPU-fastNVENC encoding from €92.59 per month (billed by the hour)
Calculate costs using the price calculator
  • FFmpeg pipelines – full control over profiles
  • Auto-scaling – Workers follow the queue
  • ABR sets – all resolutions automatically
  • S3 connection – In/Out with no traffic charges
  • 99.8% uptime SLA for ccloud³ VMs in the HA Pool – reliably hosted in German data centres
Recommended modules

The right centron products

Customers typically implement this use case using these building blocks – which can be combined and expanded at any time.

Cloud GPU
From
€92.59 / month
NVIDIA performance
  • RTX A4000 from €92.59 per month
  • Dedicated, not shared
  • No minimum term
Kubernetes
From
€29.99 / month
Container orchestration
  • AutoScaler included
  • Traffic at a fixed price
  • CI/CD-ready
S3 Object Storage
From
€5.00 / month
Scalable storage
  • S3-compatible API
  • Free traffic
  • Unlimited scalability
In a nutshell

How much does video encoding cost at Centron?

GPU-accelerated video encoding: Scale transcoding as required – pay by the hour instead of buying an encoding farm. The cornerstone of this service is Cloud GPU, starting at €92.59 per month – billed by the hour, with no minimum contract term. This is supplemented, as required, by Kubernetes and S3 object storage. The infrastructure runs in our own German data centres, which are certified to ISO 27001 on the basis of IT-Grundschutz and hold an unrestricted BSI C5:2020 Type 1 certificate. New accounts receive a €200 starting credit.

Packages and prices
Building blockPrice
Cloud GPUfrom €92.59 per month
Kubernetesfrom €29.99 per month
S3 Object Storagefrom €5.00 per month
Infrastructure & Data FAQ

Frequently asked questions

What is video encoding?

Video encoding is the process in which raw video data is compressed using various codecs and converted into a uniform, playable format.

Why should I use video encoding?

Video encoding reduces the bandwidth required and lowers storage and delivery costs. At the same time it increases compatibility, as videos in the right format can be played smoothly on a wide range of devices.

How does video encoding work in the cloud?

With cloud-based video encoding, the conversion of your video files is carried out on highly scalable servers instead of taxing your own hardware or software. You upload the material, define format and quality parameters, and processing runs flexibly and scalably in the cloud – at centron on GPU instances with results in S3 Object Storage.

How much faster is GPU encoding really?

Significantly: NVENC on RTX cards transcodes typical streams many times faster than pure CPU encodes – at predictable hourly costs. The realistic acceleration for your material is best tested on real files with the €200 starting credit.

What are codecs?

The term codec is made up of “coder” and “decoder”. A codec compresses video data (encoding) to reduce file size and decompresses it again later (decoding). Common examples are H.264, H.265, VP9 and AV1.

How much does it cost to get started?

The starting prices are deliberately low: Cloud GPU from €92.59 per month, Kubernetes from €29.99 per month and S3 Object Storage from €5.00 per month. New accounts receive a €200 starting credit valid for 60 days – you can calculate the cost of your specific configuration transparently using the price calculator.

Why should I choose centron as my cloud hosting provider?

centron offers flexible, scalable solutions designed specifically for companies that need performance and security – with current hardware, high availability from German data centres and support from experienced experts.

What advantages does centron offer for video encoding cloud hosting?

High availability with a 99.8% uptime SLA for ccloud³ VMs in the High Availability Pool according to the service specification, flexible scaling for short-term load peaks, current hardware for maximum performance, Kubernetes or root servers for individual requirements, and personal support from experienced experts.

Can this be implemented in a way that complies with the GDPR?

Yes. Encoding produces intermediate states and transcripts that inherit the personal references of the source material – they are subject to the same requirements as the original. At centron, the entire processing chain remains in Germany. The data centres are certified to ISO 27001 based on IT-Grundschutz; for ccloud³ / Managed Cloud there is an unqualified BSI C5:2020 Type 1 attestation. Details can be found in the Trust Center.

Can I rent servers from centron?

Yes, you can rent various server solutions from centron that are ideally suited to video encoding: virtual machines with GPU option and managed servers, where centron takes over operation and maintenance.

Get started for free

Sign up and receive €200 credit at centron within your first 60 days.

This promotional offer applies to new accounts only. Available exclusively to businesses.