# HTA dual cores – batch processing

Source: [https://docs.qualcomm.com/doc/80-88500-4/topic/148_HTA_230_features.html](https://docs.qualcomm.com/doc/80-88500-4/topic/148_HTA_230_features.html)

The HTA dual cores can be used to perform batch processing for multiple images using
    the Inception v3 neural network architecture.

Figure : HTA dual cores (batching example with Inception v3)
        (1)
      
      ![](data:image/png;base64,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)

- Recommended usage model for maximum throughput
- Two separate frameworks running asynchronously
- Both HTA cores working on separate inputs

Figure : HTA dual cores (batching example with Inception v3)
        (2)
      
      ![](data:image/png;base64,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)

- Only one framework, but sending a batch of two (two images at the same time) to the hardware
- Inferences occurring in parallel
- 5% throughput reduction vs. the previous figure configuration

**Parent Topic:** [Qualcomm® Hexagon™ Tensor Accelerator](https://docs.qualcomm.com/doc/80-88500-4/topic/147_HTA.html)

Last Published: Aug 18, 2023

[Previous Topic
Inception v3 performance](https://docs.qualcomm.com/bundle/publicresource/80-88500-4/topics/151_Inception_v3_performance.md) [Next Topic
Qualcomm Computer Vision SDK](https://docs.qualcomm.com/bundle/publicresource/80-88500-4/topics/152_Qualcomm_Computer_Vision_SDK.md)