By: Ronald Maas (rmaas.delete@this.wiwo.nl), August 10, 2014 7:12 pm
Room: Moderated Discussions
David Kanter (dkanter.delete@this.realworldtech.com) on August 8, 2014 11:51 pm wrote:
> Which architecture? I haven't heard of anything from any ARM-based vendor
> that would credibly challenge a Xeon in terms of performance.
>
Listen to "State of ARM and aarch64 in Fedora" by Peter Robinson (on Youtube). Around the 25:00 minute mark he mentioned on ARMv8 builds run faster compared to x86_64. ARMv8 systems are most likely Applied Micro's Mustang boards. Unfortunately I Was not able to find what hardware Red Hat is using for x86_64 builds, but assume these are fairly typical 4 core Xeons.
In the high end Intel and IBM Power8 will likely keep the lead in absolute performance above ARMv8 systems coming to the market in the next years. Agree with you that in this space difference in instruction set do not matter that much.
But most revenue is not made in the high end. So in order to successfully compete against Intel, ARMv8 systems needs to be able to provide comparable performance against Xeons in the USD 500-1000 segment at a lower TCO. And that, I believe, is achievable.
> Which architecture? I haven't heard of anything from any ARM-based vendor
> that would credibly challenge a Xeon in terms of performance.
>
Listen to "State of ARM and aarch64 in Fedora" by Peter Robinson (on Youtube). Around the 25:00 minute mark he mentioned on ARMv8 builds run faster compared to x86_64. ARMv8 systems are most likely Applied Micro's Mustang boards. Unfortunately I Was not able to find what hardware Red Hat is using for x86_64 builds, but assume these are fairly typical 4 core Xeons.
In the high end Intel and IBM Power8 will likely keep the lead in absolute performance above ARMv8 systems coming to the market in the next years. Agree with you that in this space difference in instruction set do not matter that much.
But most revenue is not made in the high end. So in order to successfully compete against Intel, ARMv8 systems needs to be able to provide comparable performance against Xeons in the USD 500-1000 segment at a lower TCO. And that, I believe, is achievable.
