Feeds:
Posts
Comments

Archive for the ‘Advanced Computing Platform’ Category

Bitcoin Network Speed 8 Times Faster than Top 500 Supercomputers Combined

Reporter: Aviva Lev-Ari, PhD, RN

 

See on Scoop.it – Cardiovascular Disease: PHARMACO-THERAPY

The bitcoin network speed estimate on bitcoinwatch.com passed 1 exaFLOPS (1,000 petaFLOPS) this week – over 8 times the combined speed of the top 500 supercomputers.

 

The bitcoin network hashrate estimate on bitcoinwatch.compassed 1 exaFLOPS (1,000 petaFLOPS) this week – over 8 times the combined speed of the top 500 supercomputers. Experts will be quick to point out that this estimate is flawed, since no FLOPS are actually used in bitcoin mining. FLOPS stands for FLoating-point Operations Per Second, and is frequently used as a standard to measure computer speed. Bitcoin mining uses an integer calculation and almost no floating-point operations, so converting bitcoin network speed to this standard is somewhat clumsy.

 

The FLOPS estimate is based on the opportunity cost of computers using their hardware for mining instead of other applications.  Miners are using their graphics cards to perform hashes instead of other FLOPS-based distributed computing. Therefore, a conversion rate of 1 hash = 12.7K FLOP is used to estimate what this hardware could be doing.

 

The estimate was created in 2011, before the production of ASIC hardware that now dominates the network. ASICS are custom designed chips that can only perform bitcoin mining calculations. The exaFLOPS estimate breaks down with ASICs, because they are not capable of floating-point operations, and therefore there is no opportunity cost associated with their use.

 

Interestingly, the estimate may still be useful for estimating how well other supercomputers and distributed networking projects would be able to mine bitcoins. Their speed is measured in FLOPS, but they also have the capability of performing the integer operations used in hashing. What would happen if the top 10 supercomputers all switched to bitcoin mining? How much would that affect the network? Lets reverse the equation, and say that they would receive 1 hash for every 12.7k FLOP.

 

The fastest computer, Sequoia, would measure at about 1.6% of the bitcoin network. Their combined speed is 48 petaFLOPS, roughly equivalent to 5% of the bitcoin network. In fact, the top 500 supercomputers have a combined speed of 12% of the bitcoin network.

 

To actually use these computers for mining, It would take more than just installing standard mining software. But lets be honest, these computers have better things to work on like curing cancer, solving global warming, and monitoring banking transactions.

See on www.thegenesisblock.com

Read Full Post »

New device can extract human DNA with full genetic data in minutes

Reporter: Aviva Lev-Ari, PhD, RN

 

See on Scoop.it – Amazing Science

Take a swab of saliva from your mouth and within minutes your DNA could be ready for analysis and genome sequencing with the help of a new device.

 

University of Washington engineers and NanoFacture, a Bellevue, Wash., company, have created a device that can extract human DNA from fluid samples in a simpler, more efficient and environmentally friendly way than conventional methods.

 

The device will give hospitals and research labs a much easier way to separate DNA from human fluid samples, which will help with genome sequencing, disease diagnosis and forensic investigations.

 

“It’s very complex to extract DNA,” said Jae-Hyun Chung, a UW associate professor of mechanical engineering who led the research. “When you think of the current procedure, the equivalent is like collecting human hairs using a construction crane.”

 

This technology aims to clear those hurdles. The small, box-shaped kit now is ready for manufacturing, then eventual distribution to hospitals and clinics. NanoFacture, a UW spinout company, signed a contract with Korean manufacturer KNR Systems last month at aceremony in Olympia, Wash.

 

The UW, led by Chung, spearheaded the research and invention of the technology, and still manages the intellectual property. Separating DNA from bodily fluids is a cumbersome process that’s become a bottleneck as scientists make advances in genome sequencing, particularly for disease prevention and treatment. The market for DNA preparation alone is about $3 billion each year.

 

Conventional methods use a centrifuge to spin and separate DNA molecules or strain them from a fluid sample with a micro-filter, but these processes take 20 to 30 minutes to complete and can require excessive toxic chemicals.

 

UW engineers designed microscopic probes that dip into a fluid sample – saliva, sputum or blood – and apply an electric field within the liquid. That draws particles to concentrate around the surface of the tiny probe. Larger particles hit the tip and swerve away, but DNA-sized molecules stick to the probe and are trapped on the surface. It takes two or three minutes to separate and purify DNA using this technology.

See on www.washington.edu

Read Full Post »

The Era of Big Data is Here: Case Studies

Reporter: Larry H. Bernstein, MD, FCAP

 

 

You are invited to attend the panel discussion
The Era of Big Data is Here: Case Studies,

hosted by the Committee on Applied and Theoretical Statistics (CATS)

of the National Research Council of the National Academies.

Thursday, October 11, 2012

3:00 p.m. – 5:00 p.m.

Room 125

National Academy of Sciences Building

2101 Constitution Ave NW, Washington, DC 20418

Speakers and topics include:

Dan Crichton

Program Manager at NASA’s Jet Propulsion Laboratory

on the roles of massive data in science and engineering

Deepak Agarwal

Director of Relevance Science at LinkedIn

on the applications of massive data in business

Gaddy Getz

Director of Cancer Genome Computational Analysis

at the Broad Institute of MIT and Harvard University

on massive data exploration in genomics

This event is free and open to the public.

Please RSVP to Michelle Schwalbe at mschwalbe@nas.edu by Tuesday, October 9.


The National Research Council established the Committee on Applied and Theoretical Statistics(CATS)
in 1978 to provide a locus of activity and concern for the statistical sciences, statistical education,
use of statistics, and issues affecting the field. CATS occupies a pivotal position in the statistical
community, providing expertise in methodology and policy formation.
 

You are receiving this one-time email because of your potential interest in this topic; you have NOT
been added to any of the National Academies listservs because of this email. To opt out from future
communications of this sort, please reply with “unsubscribe” in the subject. However, if you wish
to receive periodic updates on CATS, the Board on Mathematical Sciences and Their Applications,
or all projects and events within the Division on Engineering and Physical Sciences (DEPS), please
subscribe to  the monthly 
DEPSNews.

Read Full Post »

« Newer Posts