Improving performance of lockless data structures
As multi-core computer systems grow in number of processors and processor cores one of the largest challenges in software is to scale the performance of data structures to properly utilize the potential concurrent performance that modern hardware offers. To tap into this potential performance data structures need to be able to execute multiple operations at the same time with minimal synchronization between threads or else the hot path of the code risks stalling threads and wasting processor cycles. Below is probably the most common solution you'll find.
Read full article from Improving performance of lockless data structures
No comments:
Post a Comment