Date: Wed, 05 Feb 2014 00:55:22 -0700 From: Kevin Bowling <kevin.bowling@kev009.com> To: freebsd-hackers@freebsd.org Subject: Re: opteron a1100 arm Message-ID: <lcsqov$ui9$1@ger.gmane.org> In-Reply-To: <23B18B88-D888-46B3-99F6-905F86E20FAF@netgate.com> References: <alpine.BSF.2.00.1401311911120.2427@wojtek.tensor.gdynia.pl> <1391538649.19169.79261269.3C5F49D1@webmail.messagingengine.com> <CAFU734xXWyc_TqBJ7e4MhD2nB01BAejR_1vT9%2B_5Ar5mJncncA@mail.gmail.com> <493DEB39-C4B4-409E-B8B2-B1B11E013754@netgate.com> <60555.1391549390@critter.freebsd.dk> <23B18B88-D888-46B3-99F6-905F86E20FAF@netgate.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2/4/2014 2:42 PM, Jim Thompson wrote: > > On Feb 4, 2014, at 3:29 PM, Poul-Henning Kamp <phk@phk.freebsd.dk> wrote: > >> In message <493DEB39-C4B4-409E-B8B2-B1B11E013754@netgate.com>, Jim Thompson wri >> tes: >> >>>> No but it may well be an early reminder of the upcoming generation of >>>> powerful ARM servers that we don't want to leave unsupported. >>> >>> isn't that attractive when the 8-core, 64-bit Intel C20 >>> 00 parts are here, now, at a lower TDP >>> (20W, .vs 25W for the a1100. 22nm rocks). >> >> I very much welcome a competing 64bit CPU into the marketplace and >> will buy one myself, as soon as I can, for no other reason than to >> help break the X86 monopoly on server architecture. >> >> Monopolies are never a good thing. > > True, but I didn’t say that the chip wasn’t interesting. What I said is that it’s not that attractive (to the real market for these: micro servers). > > The dual 10Gig Ethernet and 8 SATA 3.0 ports are interesting. You won’t get that with a C2K system at 25W TDP, (4 x GigE that can run at 2.5Gbps per port, and 2 SATA 3.0 ports currently) but Intel owns IP for both, so if that becomes a differentiator for design wins, I’d expect a future variant to cover. > > But by all means, port FreeBSD to it. Perhaps it can be the long-desired “reference platform” to bring ARM into a “Tier 1” architecture status. > > Jim Where A1100 wins hands down is memory capacity, and possibly even memory bandwidth (DDR4 mentioned in the PR..). Intel's server Atom chips and even the extremely powerful Xeon E3 are quite limited by 32GB RAM (and it's also somewhat expensive vs RDIMMs). This just doesn't cut it for a ton of workloads either family of Intel CPU is otherwise well suited for. The A1100 is going to force Intel to fix that nonsense quickly or see a rapid exodus on the low end.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?lcsqov$ui9$1>