lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  PHC 
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Fri, 25 Mar 2022 22:46:54 +0800
From:   Mingbao Sun <>
To:     Sagi Grimberg <>
Cc:     Keith Busch <>, Jens Axboe <>,
        Christoph Hellwig <>,
        Chaitanya Kulkarni <>,,,
        Eric Dumazet <>,
        "David S . Miller" <>,
        Hideaki YOSHIFUJI <>,
        David Ahern <>,
        Jakub Kicinski <>,,,,,,
Subject: Re: [PATCH v2 2/3] nvme-tcp: support specifying the

Hope the 3 combinations can support the claims in the commit message.

Since for the later 2 combinations, due to packet dropping and
timeout-retransmission, the bandwidth of each TX node could suddenly
drop a few hundred MB/S. 
And on the RX node, the total bandwidth can not reach to the full link
bandwidth (which is about 6 GB/S).
In contrast, for the first combination, the bandwidth of each TX node
is stable at ~ 2GB/S.
And on the RX node, the total bandwidth reached to the full link bandwidth.
And no packet dropping occurs on the 2 switches.
This is even competitive to the performance of RDMA.

Powered by blists - more mailing lists