Q: Processing requirements for 1,000,000 photos per day with Perfectly Clear SDK
- One million photos per day, steady-state demand, spread evenly over a 24 hour period. That implies 42k photos per hour.
- Average image size is 10 megapixels.
- Applying Core (excluding noise removal) + Beautify corrections using the Perfectly Clear SDK. Note: Applying our Core corrections only will be approximately 15% processing faster.
- Basing performance off of data gathered and published by Athentech here: http://www.athentech.com/help/faq/perfectly-clear-lab-sdk-speed-performance
and verified on small test processing runs on AWS EC2 instances.
- Costs based on compute-cost only (no bandwidth, storage, load balancing, etc) with current AWS Asia-Pacific Datacenter pricing, Reserved Instance – Partial pricing model
- I/O time is included, based on relatively fast SSD’s. This totaled 13% of the overall processing time. Thus less processing capacity will be needed if the Perfectly Clear SDK is implemented in an existing image processing pipeline where these I/O tasks won’t be needed.
Using m4.2xlarge instances (8 vCPU Intel Xeon E5-2686 @2.4GHz Broadwell and 32 GB Ram), each instance would process 4 to 8 images concurrently. Lower concurrency leads to completion of individual images more quickly, while higher concurrency leads to marginally better overall throughput. Each instance would handle 3900 to 4400 images per hour. 10 systems would provide 94% to 98% utilization for approximately $2400 USD / month.
To provide on-premises processing, the same compute power as described above would be needed. Higher clock-speed CPU’s will provide near linear performance advantage. As an example, a Dell PowerEdge R330 with one Xeon E3-1240 @ 3.5GHz and 32 GB RAM should provide the processing capability of approximately 1.3 of the AWS m4.2xlarge instances. This system would retail for approximately $1200 (excluding networking, storage, OS, etc), and 8 of these would offer 91% utilization for 1,000,000 images per day, for a one-time investment of $9600 USD