ClusterInfo::process_packets handles incoming packets in a thread_pool: https://github.com/solana-labs/solana/blob/87311cce7/core/src/cluster_info.rs#L2118-L2134 However, profiling runtime shows that threads are not well utilized and a lot of the processing is done sequentially. This commit redistributes the work done in parallel. Testing on a gce cluster shows 20%+ improvement in processing gossip packets with much smaller variations. |
||
---|---|---|
.. | ||
cuda_runtime.rs | ||
lib.rs | ||
packet.rs | ||
perf_libs.rs | ||
recycler.rs | ||
recycler_cache.rs | ||
sigverify.rs | ||
test_tx.rs |