How to estimate Network Speed for Guinness World Record

kano

legendary

Activity: 4634

Merit: 1851

Linux since 1997 RedHat 4

Quote from: EricSU on October 14, 2011, 07:49:06 PM

I don't think bitcoin will qualify as a "distributed computing network".
No work is distributed between bitcoin clients.
Each client works individually on its own block, its not a shared effort on the same job.

Pooled mining servers like deepbit is a distributed computing network.
Because the pool tells all nodes to work on the same block.

Actually:
They are all trying to find the 'next' block.
Although the exact contents of each attempted solution is slightly different (every single hash done is slightly different), the effect of what everyone attempts is the same:

They are all trying to confirm outstanding transactions if any are available.

Even at a single pool, the difference you stated is still there.
Everyone is trying to find the 'next' block but with a different nonce value and with different transactions as more become available.
Each getwork may include new transactions for the same block
i.e. each getwork is different as with a similar difference as comparing 2 getworks from 2 different pools at the same time.

phelix

legendary

Activity: 1708

Merit: 1020

Quote from: EricSU on October 14, 2011, 07:49:06 PM

I don't think bitcoin will qualify as a "distributed computing network".
No work is distributed between bitcoin clients.
Each client works individually on its own block, its not a shared effort on the same job.

Pooled mining servers like deepbit is a distributed computing network.
Because the pool tells all nodes to work on the same block.

they are all trying to find a block. is that very different from folding@home?

EricSU

newbie

Activity: 25

Merit: 0

I don't think bitcoin will qualify as a "distributed computing network".
No work is distributed between bitcoin clients.
Each client works individually on its own block, its not a shared effort on the same job.

Pooled mining servers like deepbit is a distributed computing network.
Because the pool tells all nodes to work on the same block.

phelix

legendary

Activity: 1708

Merit: 1020

Quote from: ptshamrock on October 10, 2011, 08:09:47 AM

thx phelix for managing this here . I like it very much!!

A Fan !

thanks!

to really officially get the record we will have to bother Guinness some more...

ptshamrock

hero member

Activity: 484

Merit: 500

thx phelix for managing this here . I like it very much!!

A Fan !

phelix

legendary

Activity: 1708

Merit: 1020

Quote from: kano on October 01, 2011, 08:32:33 PM

Just in case you do actually send this to them ...
The PoW is actually a double sha256 (for that OPs/Hash number)

Also, it's very similar to sha256(sha256(data)) but not exactly
(part of the start isn't calculated every time and the end is cut short thus the actual value of sha256(sha256(data)) isn't completed)

yeah, Phateus explains it here: https://bitcointalksearch.org/topic/m.550288

he has releasd the most optimized public kernel so far.

kano

legendary

Activity: 4634

Merit: 1851

Linux since 1997 RedHat 4

Just in case you do actually send this to them ...
The PoW is actually a double sha256 (for that OPs/Hash number)

Also, it's very similar to sha256(sha256(data)) but not exactly
(part of the start isn't calculated every time and the end is cut short thus the actual value of sha256(sha256(data)) isn't completed)

phelix

legendary

Activity: 1708

Merit: 1020

Quote from: maaku on September 30, 2011, 12:12:06 PM

I'm not sure why there has to be voodoo involved. Most of the bitcoin network is GPU-based (at the moment). Modern GPUs can perform true floating-point operations. So get the most popular GPUs in use, get their IOPS and FLOPS ratings, and use the median ratio for calculating the Guinnes world record value.

AMD uses a factor of 2 for their Radeon cards

we do not know how much of the Hashrate is generated from FPGAs for example which may not be able to do any FLOPs at all. I think non GPU computing power is still neglectable, though.

maaku

legendary

Activity: 905

Merit: 1012

I'm not sure why there has to be voodoo involved. Most of the bitcoin network is GPU-based (at the moment). Modern GPUs can perform true floating-point operations. So get the most popular GPUs in use, get their IOPS and FLOPS ratings, and use the median ratio for calculating the Guinnes world record value.

phelix

legendary

Activity: 1708

Merit: 1020

#python script for you to play with

D = 1689334.404597 # difficulty
t = 641.79 # time per block
h = D* 2**48 / 0xffff # average hashes per block
H = h / t # network hashes per sec
print "network hashes/sec H:", H

a = 3385 # ALU OPs / Hash (more on 6xxx cards)
i = a * H # INTOPs / sec
print "network PetaINTOPs/sec:", i / 10**15

f = i * 2 # maybe 1.0 or 1.5 to be more conservative and make up for FPGAs
print "network would-be PetaFLOPs/s (=voodoo):", f / 10**15

#output
##network hashes/sec H: 1.13054842451e+13
##network PetaINTOPs/sec: 38.2690641697
##network would-be PetaFLOPs/s (=voodoo): 76.5381283393

phelix

legendary

Activity: 1708

Merit: 1020

turns out a gpu does a minimum of 3375 integer operations per hash

https://bitcointalksearch.org/topic/m.550288

phelix

legendary

Activity: 1708

Merit: 1020

Quote from: BurtW on September 29, 2011, 08:19:19 AM

This is exactly how bitcoinwatch estimates FLOPS. I am not saying it is an acurate way to do the estimate, all I am saying is this is how they do it for what it is worth and for comparison to your other methods. Many here among us question the valididty of this estimation method.

The page simply uses the following assumptions/estimates:

1 INTOP = 2 FLOP
1 hash = 6.35K INTOP
1 hash = 12.7K FLOP

So the hashrate in TeraFLOP/s is simply 12.7 times the hashrate in Gigahashes/s.

As an example: 11,558.55 Gigahashs/s * 12.7 TeraFLOP/Gigahash = 146,794 TeraFLOP/s = 146 PetaFLOP/s

yes, somebody said so in an older post (link above).

Quote

1 hash = 6.35K INTOP

this is too high in my opinion. ArtForz threw it into the ring for x86 INTOPs.

luckily we are in the comfortable situation that however conservative we make the comparison we still are the fastest Grin

BurtW

legendary

Activity: 2646

Merit: 1137

All paid signature campaigns should be banned.

This is exactly how bitcoinwatch estimates FLOPS. I am not saying it is an acurate way to do the estimate, all I am saying is this is how they do it for what it is worth and for comparison to your other methods. Many here among us question the valididty of this estimation method.

The page simply uses the following assumptions/estimates:

1 INTOP = 2 FLOP
1 hash = 6.35K INTOP
1 hash = 12.7K FLOP

So the hashrate in TeraFLOP/s is simply 12.7 times the hashrate in Gigahashes/s.

As an example: 11,558.55 Gigahashs/s * 12.7 TeraFLOP/Gigahash = 146,794 TeraFLOP/s = 146 PetaFLOP/s

phelix

legendary

Activity: 1708

Merit: 1020

edit with new # of intops / hash (2011-09-30)

from blockexplorer:
147282    64afcd14e7...    2011-09-28 19:06:58    93    8775.96044812    51.365
147263    145d985b3f...    2011-09-28 15:43:44    13    251.70410514    6.011

--> 19 new Blocks in 12194 Secs

t ~= 641.79 secs / block (10.7 min)

D = 1689334.404597

The expected number of hashes we need to calculate to find a block with difficulty D is [...]
D * 2**48 / 0xffff
(https://en.bitcoin.it/wiki/Difficulty)

= h = 7255746733660647.0

H = h / t

--> Network Hashrate
H ~= 11305484245096.756 Hashes / sec

Int-Ops / Hash

a = 3385 ALU OPs / Hash (for 5xxx architecture; 6xxx: 4220 ---> we go for a conservative value)
(https://bitcointalksearch.org/topic/m.550288)

IntOps / Sec:
i = a * H

i = 38.2690641697 PetaIntOps / s

For comparison:
for Radeon Cards AMD says 1 32Bit-IntOp = 2 16Bit FLOPs (though it may only make limited Sense)

f = i * 2 = 76.5381283393 PetaFLOPs/s

off Bitcoinwatch by a factor of two because Bitcoinwatch uses a different number of INTOPs / hash (x86).

Bitcoinwatch: Network Hashrate PetaFLOPs/s    153.06

Anyway still more than 18 times faster than folding@home, and also much faster than all the others like SETI, BOINC, etc.

Hereby I declare BITCOIN the MOST POWERFUL DISTRIBUTED COMPUTING NETWORK on earth.

phelix

legendary

Activity: 1708

Merit: 1020

Quote from: dishwara on September 29, 2011, 06:39:14 AM

We have to send letter to Guinness that our bitcoin network has highest hash power in integer calculation & ask them to record it.
Instead of just Network speed, we have to tell them that we have a new record which can be only calculated in integer, so divide the network speed record in to TWO categories one using FLOPS & other using INTEGER & ask them to record bitcoin network in INTEGER.
If enough people ask Guinness, they will surely consider it.

hmm maybe you are right.

also we should just claim the record until someone can prove the opposite. it is pretty obvious to me that bitcoin is the fastest network. I will post a new calculation soon.

dishwara

legendary

Activity: 1855

Merit: 1016

We have to send letter to Guinness that our bitcoin network has highest hash power in integer calculation & ask them to record it.
Instead of just Network speed, we have to tell them that we have a new record which can be only calculated in integer, so divide the network speed record in to TWO categories one using FLOPS & other using INTEGER & ask them to record bitcoin network in INTEGER.
If enough people ask Guinness, they will surely consider it.

phelix

legendary

Activity: 1708

Merit: 1020

folding@home stats:
http://fah-web.stanford.edu/cgi-bin/main.py?qtype=osstats

foling@home flop faq:
http://folding.stanford.edu/English/FAQ-flops

phelix

legendary

Activity: 1708

Merit: 1020

Quote from: wareen on August 29, 2011, 01:21:41 PM

Quote from: kjj on August 29, 2011, 01:12:27 PM

I'm pretty sure the Guinness people aren't interested in what we could do, but only in what we actually do do. The right approach would probably be to ask them to consider adding a category that we do fit into.

Well, just because some theoretical supercomputer might usually work on integer problems doesn't make it less a supercomputer worthy of Guinness' attention. I'd say its the hardware that counts, not the software usually running on it.

I think the proper way would be to explain the situation, give them the best estimates of the hardware (and a FLOPs value) we can come up with and let them decide what to do.

+1 that's exactly how I see it.

folding@home who currently own the title also mostly consist of the very same GPUs.

kjj

legendary

Activity: 1302

Merit: 1026

Quote from: wareen on August 29, 2011, 01:21:41 PM

Quote from: kjj on August 29, 2011, 01:12:27 PM

I'm pretty sure the Guinness people aren't interested in what we could do, but only in what we actually do do. The right approach would probably be to ask them to consider adding a category that we do fit into.

Well, just because some theoretical supercomputer might usually work on integer problems doesn't make it less a supercomputer worthy of Guinness' attention. I'd say its the hardware that counts, not the software usually running on it.

I think the proper way would be to explain the situation, give them the best estimates of the hardware (and a FLOPs value) we can come up with and let them decide what to do.

Just don't be disappointed. Floating point operations have been an implicit part of the definition of a supercomputer for like 45 years now.

wareen

legendary

Activity: 910

Merit: 1001

Revolutionizing Brokerage of Personal Data

Quote from: kjj on August 29, 2011, 01:12:27 PM

I'm pretty sure the Guinness people aren't interested in what we could do, but only in what we actually do do. The right approach would probably be to ask them to consider adding a category that we do fit into.

Well, just because some theoretical supercomputer might usually work on integer problems doesn't make it less a supercomputer worthy of Guinness' attention. I'd say its the hardware that counts, not the software usually running on it.

I think the proper way would be to explain the situation, give them the best estimates of the hardware (and a FLOPs value) we can come up with and let them decide what to do.

Topic: How to estimate Network Speed for Guinness World Record (Read 5665 times)