bblazei
commented
2 years ago Not a direct answer to your question but I think you're holding gRPC incorrectly:
https://www.mpi-hd.mpg.de/personalhomes/fwerner/research/2021/09/grpc-for-ipc/
Puts 99th percentile of same node separate core at 142 microseconds. Also sounds like bidirectional streaming can bring that number down more.
"10s of milliseconds" gets you to the moon and back.
On Thu, Dec 16, 2021, 4:20 AM kasajian @.***> wrote:
I've been looking for this info everywhere and can't seem to find it. Looking to see what the overhead is of an IPC call. If one process makes a call to another process, how much time takes for a round-trip call. for gRPC I'm measuring in the tens of milliseconds. seems excessive. what transport does zeromq.net use for IPC. No off-node networking. This is only for on-node IPC. Looks like from the docs we have the same richness as gRPC.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/zeromq/netmq/issues/996, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAREBWSDM3GV6THAHAY64JLURGVMVANCNFSM5KFYUSRQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
I've been looking for this info everywhere and can't seem to find it. Looking to see what the overhead is of an IPC call. If one process makes a call to another process, how much time takes for a round-trip call. for gRPC I'm measuring in the tens of milliseconds. seems excessive. what transport does zeromq.net use for IPC. No off-node networking. This is only for on-node IPC. Looks like from the docs we have the same richness as gRPC.