Performance degradation in the Marshaller Export functionality. The root cause of the issue was identified in the GridHandleTable class used inside OptimizedMarshaller, where the internal buffers would grow in size but not shrink back, leading to increased CPU consumption over time.

The fix involved optimizing the implementation of GridHandleTable by removing unnecessary operations and improving memory management. This resulted in improved performance and prevented the excessive growth of internal buffers.

5 2 - 1 Feb. 8, 2024, 12:47 PM


Launch on Chameleon

Launching this artifact will open it within Chameleon’s shared Jupyter experiment environment, which is accessible to all Chameleon users with an active allocation.

Download Archive

Download an archive containing the files of this artifact.

Version Stats

5 2 -