From: Srinivas Reddi <srinivas.reddi@aricent.com>
To: "dev@dpdk.org" <dev@dpdk.org>,
Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Subject: [dpdk-dev] MEMNIC [How can we achieve max through put using memnic interfaces]
Date: Mon, 21 Jul 2014 15:15:17 +0000 [thread overview]
Message-ID: <9918fdb09c2c4dd5a4fb40a5e6a04968@GURMBXV02.AD.ARICENT.COM> (raw)
Hi,
I wrote a test application using MEMNIC interfaces for inter VM switching .[VM1 --> host --- >VM2 ]
I send uni-directional traffic from traffic generator at vm1 to the host .
And in the host my application switches traffic coming from vm1 to vm2.
I have used dpdk-1.7 and memnic -1.2 . and librte_pmd_memnic_copy.so to bind memnic interfaces to dpdk application [my traffic generator app].
I observed a maximum throughput of around 470Mbps only [with packet size 1400 Bytes,21,000 packets per second ] ..
Is there any best method to improve throughput using MEMNIC interfaces ..
If I increase shared memory area shown below .. does it servers my need .
If I want around 4 to 5 Gbps through put with unidirectional flow from VM1 to VM2 through the host ..
How much shared memory area shall I take per port ..
What might be the memory alignment ..
/*
* Shared memory area mapping
* +------------------+
* | Header Area 1MB |
* +------------------+
* | Up to VM 7MB |
* +------------------+
* | Padding 1MB |
* +------------------+
* | Down to host 7MB |
* +------------------+
*/
struct memnic_area {
union {
struct memnic_header hdr;
char hdr_pad[1024 * 1024];
};
union {
struct memnic_data up;
char up_pad[7 * 1024 * 1024];
};
char blank[1024 * 1024];
union {
struct memnic_data down;
char down_pad[7 * 1024 * 1024];
};
};
Thanks & Regards,
Srinivas.
"DISCLAIMER: This message is proprietary to Aricent and is intended solely for the use of the individual to whom it is addressed. It may contain privileged or confidential information and should not be circulated or used for any purpose other than for what it is intended. If you have received this message in error, please notify the originator immediately. If you are not the intended recipient, you are notified that you are strictly prohibited from using, copying, altering, or disclosing the contents of this message. Aricent accepts no responsibility for loss or damage arising from the use of the information transmitted by this email including damage from virus."
reply other threads:[~2014-07-21 15:14 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=9918fdb09c2c4dd5a4fb40a5e6a04968@GURMBXV02.AD.ARICENT.COM \
--to=srinivas.reddi@aricent.com \
--cc=dev@dpdk.org \
--cc=h-shimamoto@ct.jp.nec.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).