From mboxrd@z Thu Jan  1 00:00:00 1970
MIME-Version: 1.0
In-Reply-To: <CAOw7k5ijD9K2=DMJQ7orO5o=Kzw5KGie-nPfn+Sgk3z8PXdJtA@mail.gmail.com>
References: <F022808EFD5F0B8A438B332081972DD2@felloff.net>
	<CAGGHmKGPVcmq2XkYe21rmPMf2JjdYbUn4GgjanMrQEWg0TW41A@mail.gmail.com>
	<B48DC5FB-1125-48A4-A685-BBB7854F305C@bitblocks.com>
	<7aa20c07322147e6@orthanc.ca>
	<20181011230430.8ec148a3cb2a4d95180a4ad2@eigenstate.org>
	<CAOw7k5jAHW_aDZmQdqVsNeXZwsiRvWRAhcYYfWAiPjEREKcxaQ@mail.gmail.com>
	<CAFSF3XM4=g9q4sj1Om9EAjCBSOEjRga2X0rjfjeF4KM1SEbATw@mail.gmail.com>
	<4A1686D0-C80C-417F-A3D6-3F9EA327D35F@lsub.org>
	<CAFSF3XMgB=QLrtqKpxq7MBVAm2ZOKTK7JxrMWWk24HcYZKmeXg@mail.gmail.com>
	<FA46EAF2-2344-4DA0-8A56-7B85CD4C4826@lsub.org>
	<CAFSF3XOWCSMM711i+mTw=qSthxE6-bTOZj4vFBHrC0OHQCEy7Q@mail.gmail.com>
	<CAOw7k5ijD9K2=DMJQ7orO5o=Kzw5KGie-nPfn+Sgk3z8PXdJtA@mail.gmail.com>
From: hiro <23hiro@gmail.com>
Date: Mon, 15 Oct 2018 19:29:33 +0200
Message-ID: <CAFSF3XPTnLYh-tQGYqk5hmpR1dVZ7=omo6scbsmocru54HubAg@mail.gmail.com>
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Content-Type: text/plain; charset="UTF-8"
Subject: Re: [9fans] zero copy & 9p (was Re: PDP11 (Was: Re: what heavy
	negativity!)
Topicbox-Message-UUID: edcd05bc-ead9-11e9-9d60-3106f5b1d025

> Also, NUMA effects are more important in practice on big multicores. Some
> of the off-chip delays are brutal.

yeah, we've been talking about this on #cat-v. even inside one CPU
package amd puts multiple dies nowadays, and the cross-die cpu cache
access delays are approaching the same dimensions as memory-access!

also on each die, they have what they call ccx (cpu complex),
groupings of 4 cores, which are connected much faster internally than
towards the other ccx