In this video episode I discussed a possibility that you can offload OpenZFS lossless compression to any dedicated lossless compression hardware accelerator cards. Usually such hardware can be used for various applications such as generic file compression, NAS storage stack, WAN Optimization, VPN stack and so on. At high bandwidth doing any lossless compression on a CPU (a generic lossless compression library in kernel/user-space API uses nothing but CPU/Memory resources) is a huge overload, which we overcome by offloading on to a dedicated high-bandwidth hardware.
There are many such accelerator cards available in the market. Some cards does both compression as well high-speed encryption thus enables us to overcome this CPU (and RAM) bound latency overheads.
For example: Here is a general purpose AHA AHA363PCIE0301G 5Gbs GZIP Compression/Decompression Accelerator Card. Typical applications (or use-cases) of this card could be Storage Arrays, Load Balancers, WAN Optimization, Web Servers, Data Analytics, etc.
Intel FPGA PAC D5005 High-end Drop-in Accelerator: Here is yet another but general purpose FPGA Accelerator card can be used for processing high-bandwidth Network and Storage Data Processing (hardware offload). Read the complete article here.