Compression and decompress performance for mongodb is very important, a lot of mongodb deployment enabled compress/decompress. Meanwhile, Intel made an announcement on its accelerators called (IAA - in memory analytics accelerator) on next generation Xeon Scalable processor (code name Sapphire Rapids), _such as very high throughput compression and decompression combined with primitive analytic functions, as well as to provide highly-optimized SW fallback on other Intel CPUs.  Intel demonstrated IAA performance on this link: https://www.servethehome.com/intel-xeon-sapphire-rapids-shows-built-in-accelerators-at-innovation-2022/

Here we've made an prototype based on Intel IAA to speedup compress/decompress in MongoDB, and proved good value prop when test with YCSB, much better compression compared with Snappy meanwhile better performance for insert (compress) and queries (decompress). 

we want to upstream our code to Wiredtiger. pls feedback. 

we could provide remote server for you to reproduce the performance. pls contact with us.