# FSISUS15: What is your testing process for workloads that require floating point precision?
<a name="fsisus15"></a>

## FSISUS15-BP01 Minimize the bit count while maintaining precision
<a name="fsisus15-bp01-minimize-the-bit-count-while-maintaining-precision"></a>

### Prescriptive guidance
<a name="prescriptive-guidance-23"></a>

 Floating point precision is a way to represent real numbers in a finite binary format. It stores a number in a fixed-width field with the intent to reduce the memory bandwidth and storage requirements compared to double-precision arithmetic results. Although double-precision can sometimes lead to more accurate results, single-precision calculations can be faster and thus 

 reduce overall energy consumption for particular workloads. Determine which of your workloads is suitable for use of floating-point accuracy, performance, and efficiency. Consider testing with a cluster of instances to see how well it performs at scale. 

### Implementation guidance:
<a name="implementation-guidance-1"></a>
+  For intensive financial simulations and calculations, test the number of bits that are required to achieve your floating point precision and consider reducing number of bits by selecting different floating-point formats, including bfloat16, that's supported by AWS Graviton. 
+  Using floating point [Quantization](https://aws.amazon.com/blogs/machine-learning/reduce-ml-inference-costs-on-amazon-sagemaker-with-hardware-and-software-acceleration/), you can represent numbers using lower bit-count integers or floating point numbers without incurring a significant loss in accuracy. Specifically, you can reduce resource usage by replacing the parameters in your workload with (1) half-precision (16 bit), (2) bfloat16 (16 bit, but the same dynamic range as 32 bit), or 8-bit integers instead of the usual single-precision floating-point (32 bit) values. 
+  **Service recommendations:** Use the following services to achieve your goal. 
  +  [AWS Batch](https://aws.amazon.com/batch/) 
  +  [AWS Parallel Cluster](https://aws.amazon.com/hpc/parallelcluster/) 
  +  [Graviton3](https://aws.amazon.com/about-aws/whats-new/2022/05/amazon-ec2-c7g-instances-powered-aws-graviton3-processors/) 
+  Test generative AI models with reduced precision (quantization) to maintain accuracy while reducing resource consumption. 
+  Validate generative AI model performance with different floating-point precisions. 
+  Use mixed-precision training for generative AI models to optimize resource usage.