Benchmark between warboy and coral.ai for CCTV

Furiosa Runtime
1

Please advise about benchmark between warboy and coral.ai for CCTV application.

2

Dear Tom,
I think we could help you better if you can share the benchmark scenario in detail.

Since no specific product was specified, I will assume the SoM type for coral.ai., and compare the HW specs between the two models.

Warboy has a significantly higher peak performance (64 TOPS) and memory capacity(16GB) compared to Coral’s 4 TOPS and 1GB. Warboy would likely be moresuitable for intensive ML workloads, like real-time video analytics,

Coral System-on-Module (SoM) Specs:

  • CPU: NXP i.MX 8M SoC (quad Cortex-A53, Cortex-M4F)
  • ML accelerator: Google Edge TPU coprocessor, 4 TOPS (int8); 2 TOPS per watt
  • RAM: 1 GB LPDDR4 (options for 2 GB or 4 GB coming soon)
  • Flash Memory: 8 GB eMMC
  • Wireless: Wi-Fi 2x2 MIMO (802.11b/g/n/ac 2.4/5GHz) and Bluetooth 4.2
  • Dimensions: 48mm x 40mm x 5mm
  • Availability: Various countries globally

Warboy Specs (from what can be inferred from the second image):

  • Peak Performance: 64 TOPS
  • On-chip SRAM: 32 MB
  • Host Interface: PCIe Gen4 8-lane
  • Form Factor: Full-Height Half-Length, also available in Half-Height Half-Length
  • Thermal Solution: Passive and Active Fan
  • TDP: 40 - 60W (Configurable)
  • Operating Temperature: 0 - 50°C
  • Clock Speed: 2.0 GHz
  • DDR Speed: 4266 Mbps
  • Memory Type: LPDDR4X
  • Memory Size: 16 GB (max. 32 GB)
  • Peak Memory Bandwidth: 66 GB/s