# Optimising Temporary Memory Usage for Convolutions and Matmuls on the IPU This directory contains the supporting code and tests for the [technical note on setting the available memory proportion for convolutions and matmuls](https://docs.graphcore.ai/projects/available-memory/en/latest/). Further information on memory optimisation can be found in the technical note [Memory and Performance Optimisation on the IPU](https://docs.graphcore.ai/projects/memory-performance-optimisation/en/latest/common-memory-optimisations.html).