[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CALW8-7+3gK6QQujmOc=1-VngxfGQQ06ukypU+AOA6D8Zw7WOgQ@mail.gmail.com>
Date: Fri, 3 Apr 2015 14:34:24 +0200
From: Dmitry Khovratovich <khovratovich@...il.com>
To: "discussions@...sword-hashing.net" <discussions@...sword-hashing.net>
Subject: Re: [PHC] Panel: Please require the finalists to help with benchmarks
On Fri, Apr 3, 2015 at 2:01 PM, Solar Designer <solar@...nwall.com> wrote:
> On Fri, Apr 03, 2015 at 12:59:53PM +0200, Dmitry Khovratovich wrote:
>> We could try to develop several typical scenarios for benchmarking.
>
> Agreed.
>
>> Maybe people from industry could contribute with usecases.
>>
>> For example:
>> Scenario 1 (cryptocurrency mining on x86 desktop):
>> maximum time: 1 second
>> maximum memory: 4 GB
>> maximum threads: unlimited
>
> I'm not an expert in cryptocurrency issues, but from what I heard so far
> the maximum time per hash computation is a lot lower, as needed to allow
> for fast verification of transactions. This is what keeps them at
> around 2 MB currently. Maybe an expert would chime in.
>
The things are a bit different. Transactions are hashed and
ECC-signed, and indeed they must be verified fast. Still, an ECC
signature check takes about 1/1000-th of a second, and the current
rate of 700 transaction per block means that the total verification
takes almost a second already.
However, mining is different, you do not hash transactions there, you
just hash a singe nonce. Spending 1 second for block hashing means
that the total block verification would take 2 seconds instead of
current 1. Not a big deal, blocks appear every 10 minutes (in Bitcoin)
anyway.
--
Best regards,
Dmitry Khovratovich
Powered by blists - more mailing lists