Snow Leopard & Distributed computing

[t0mat0]

Just wanted to set a thread up regarding Snow Leopard, and distributive computing. With Snow Leopards planned features, and theoretical limits, the potential for using Snow Leopard as an OS for distributed computing is huge.

Potentially, if the software came about, users with Snow Leopard based Macs could pool their resources and outdo efforts such as HECTOR
http://www.guardian.co.uk/technology/2008/jan/02/computing.climatechange

Basically, the communal potential is larger than most of the supercomputers, but the latent potential isn't used that much, except for projects such as Folding at home. folding.stanford.edu/
It will be very interesting to see the effects of both Snow Leopard on normal machines, on machines specifically designed to maximise power, and see that in comparison to say a more Road Runner esque Future Cell chip based supercomputer.

Just a stub for now, i'll flesh this post out with information - consider it a placeholder

 

[twoodcc]

i would love to see the distributed computing used more - and more easily used - for people that have a few macs at home - but aren't great with grid computing

 

[Apple Ink]

Since Im not a brainer in this stuff.... will the current 802.11n standard be able to sustain the technology to its full potential?

Sorry... completely stupid post.... read below!

 

[Spikeanator6982]

Since Im not a brainer in this stuff.... will the current 802.11n standard be able to sustain the technology to its full potential?

You've lost me with that question...What are you talking about?

 

[t0mat0]

From wiki:
Grid computing is a style of distributed computing whereby a "virtual supercomputer" is composed of a cluster of networked, loosely-coupled computers, acting in concert to perform very large tasks. This technology has been applied to computationally-intensive scientific, mathematical, and academic problems through volunteer computing, and it is used in commercial enterprises for such diverse applications as drug discovery, economic forecasting, seismic analysis, and back-office data processing in support of e-commerce and web services.

What distinguishes grid computing from typical cluster computing systems is that grids tend to be more loosely coupled, heterogeneous, and geographically dispersed. Also, while a computing grid may be dedicated to a specialized application, it is often constructed with the aid of general purpose grid software libraries and middleware.

Distributed computing (e.g. Folding@home ) deals with hardware and software systems containing more than one processing element or storage element, concurrent processes, or multiple programs, running under a loosely or tightly controlled regime.

In distributed computing a program is split up into parts that run simultaneously on multiple computers communicating over a network. Distributed computing is a form of parallel computing, but parallel computing is most commonly used to describe program parts running simultaneously on multiple processors in the same computer. Both types of processing require dividing a program into parts that can run simultaneously, but distributed programs often must deal with heterogeneous environments, network links of varying latencies, and unpredictable failures in the network or the computers.


Don't see where 802.11n fits in. The Snow Leopard thread has a bit more flesh on now, but it might be interesting to keep this thread concurrently, to look more into the distributed/supercomputer uses of Macs using Snow Leopard.

 

[Apple Ink]

From wiki:
Grid computing is a style of distributed computing whereby a "virtual supercomputer" is composed of a cluster of networked, loosely-coupled computers, acting in concert to perform very large tasks. This technology has been applied to computationally-intensive scientific, mathematical, and academic problems through volunteer computing, and it is used in commercial enterprises for such diverse applications as drug discovery, economic forecasting, seismic analysis, and back-office data processing in support of e-commerce and web services.

What distinguishes grid computing from typical cluster computing systems is that grids tend to be more loosely coupled, heterogeneous, and geographically dispersed. Also, while a computing grid may be dedicated to a specialized application, it is often constructed with the aid of general purpose grid software libraries and middleware.

Distributed computing (e.g. Folding@home ) deals with hardware and software systems containing more than one processing element or storage element, concurrent processes, or multiple programs, running under a loosely or tightly controlled regime.

In distributed computing a program is split up into parts that run simultaneously on multiple computers communicating over a network. Distributed computing is a form of parallel computing, but parallel computing is most commonly used to describe program parts running simultaneously on multiple processors in the same computer. Both types of processing require dividing a program into parts that can run simultaneously, but distributed programs often must deal with heterogeneous environments, network links of varying latencies, and unpredictable failures in the network or the computers.


Don't see where 802.11n fits in. The Snow Leopard thread has a bit more flesh on now, but it might be interesting to keep this thread concurrently, to look more into the distributed/supercomputer uses of Macs using Snow Leopard.

Really really sorry.... My bad I mixed up both grid and distributed....

Seems like I should've talked about CUDA and grandcentral instead Im talking about 802.11n...

 

[Muncher]

You know, it's interesting. If Apple had a service where people connected to a network of computers where people could "donate" (when they don't need it) computing power to others who were maxing out the power they already had. Most people don't use their computers to their full potential, so the network would open up a lot of wasted power.

 

[t0mat0]

Listening to a recent tech / science podcast, it seems that in a way, there seemed to be a possibility that XGrid is about to get a lot more powerful come Snow Leopard.

With the ability to set up an XGrid without needing a Server copy, and GUIs available, it seems that the GPGPU aspect, along with the effect of easily being able to find Macs to recruit for a project seems like Apple could if they wanted to, market easy to set up distributed computing that a prosumer could set up.

With the raw power of GPU cards about to kick in, and with Nehalem Mac Pros and XServes -could Apple make it easier for users to farm out their large loosely coupled work to their own Xgrids?