DPDK usage discussions
 help / color / mirror / Atom feed
* [dpdk-users] The latency of dpdk is more than 400us, something wrong?
@ 2018-05-09  2:57 shengxiao qian
  0 siblings, 0 replies; only message in thread
From: shengxiao qian @ 2018-05-09  2:57 UTC (permalink / raw)
  To: users

NIC: Intel 82599ES
DPDK version: 17.08.1


My test model is like this:
 Host A send request packet to host B,when host B  receive the request
packet,host B send replay packet to host A。Then host A send next request to
host B ...

When the request packet size is less than or equal to  128 bytes, it works
fine. the round trip latency is less than 12us. But if the packet size is
more than 128 bytes(even 129bytes),The latency is more than 400us。

I find that  most of the time consumed in the receiving package.

In my application, I call the rte_eth_tx_burst()  about every 1us.

What may be the cause? The NIC or dpdk driver receive packet made some
batch?

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2018-05-09  2:57 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-05-09  2:57 [dpdk-users] The latency of dpdk is more than 400us, something wrong? shengxiao qian

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).