Follow Slashdot blog updates by subscribing to our blog RSS feed


Forgot your password?

Comment Re:Intel's trolling us (Score 1) 310

It's not as simple as how many flops can you do. The PowerPC Top500 are created specifically certain types of workloads and are designed by to ran at 100% 24/7. The idle power consumption of those platforms is ridiculous. Super computers are nothing like computers in normal datacenters. Datacenters have large swings of computational usage and do a lot of VMs. Super computers have constant load and run bare-metal for the most part with highly optimized OS's with kernel schedulers that schedule time slices in quantums of minutes instead of milliseconds.

And yes, AMD has great peak flops and can compete with Intel in this area, but AMD's design is biased more like a GPU than a CPU and takes a big hit any time you need to do thread synchronization. Higher latency to flush a cache-line, which is highly detrimental to synchronized performance(Amdahl's law). AMD wanted a large exclusive cache instead of a smaller low latency inclusive cache. ARM is great if you only need a bunch of weak CPU cores. Below 1Ghz, ARM is slightly ahead of Intel. Around 1ghz, ARM runs into efficiency issues and suddenly they draw more power than Intel and only similar performance at best. ARM does beat Intel in some niche workloads, but the same can be said about pretty much any CPU.

Comment Re:With AMD out of the way Intel can F*** us. (Score 2) 310

Intel actively sources many of their rare minerals from companies that need a fully audited conflict-free African areas. Unlike many of the other big companies that source from the cheapest, which often is in areas that have massive conflict issues. Intel's new requirements has actually eliminated conflicts in some areas because the money dried up until the warlords moved out, then money came back in. It has had a very positive affect by bringing in much needed money but keeping warlords out of the mix.

Comment Re:Defining "Progress"... (Score 1) 310

C# can be quite decent for performance with proper design and optimizations. To get max performance, you need to stop allocating new objects and start using more structs and buffers. C# also makes threading an async simple. Although, I'm not sure it would be able to keep up with 2GiB/s. That is a lot of IO. I depends on what you're doing. Even async could have issues with 2GiB/s because async uses tasks, which are new objects when a method does not return synchronously.

Comment Re:Intel's trolling us (Score 1) 310

You seem to have no idea how much Microsoft R&D has contributed to many aspect of technology, especially around computers. A lot of modern GPUs were made possible because Microsoft had many kinds of exotic custom hardware architectures created, then create several custom OS kernels with completely different designs from mainstream, just to see what the best way forward would be. This allowed AMD, Intel, and Nvidia to prototype many ideas.

Comment Re:Intel's trolling us (Score 1) 310

Other than Intel being faster and consuming less power. Especially on high end servers. My cousin had to purchase 4 $60k servers, two Intel builds and two AMD builds. Ran them all through a slew of tests for both Linux and Solaris, everything as optimized as possible. Not only did Intel win in nearly every benchmark, but it consumed nearly 1/2 the power. This was important because the mains was only able to handle 10 megawatts and he was able to get their electrician to jimmy rig another 2 megawatts, but they still had several $500,000 racks turned off. You get to do silly things when you have your own 150 megawatt power plant that doesn't charge you any money. State funded University research center.

Comment 80/20 (Score 3, Interesting) 127

Many look at a technology and say "it's good enough, it does 80% of what I need". Then they cobble together 9 other technologies the same way and you're left with 0.8^10 "enough", leaving you fighting fires from the lack of custom configuration that you need. Technical debt is multiplicative with other technical debt.

For every 1 person that reinvents the wheel, 9 others use an existing wheel for the wrong job or misconfigure the wheel because they don't understand their problem well enough. If you truly understand your current issue, you're smart enough to create a solution. Every time someone treats a tool like a black box of magic, looking at you programmers blindly using libraries without understanding how they work, it's because they don't understand the problem they're trying to solve.

P.S. Understanding what something is doing does not mean you know the exact details of the implementation.

Comment Re: Oblig (Score 1) 153

150 isn't that great. I get 1Gb bursts from most websites which would cause ping spikes and jitter if it wasn't for my ISP using an AQM. It's not how slow the Internet is, it's how much faster it could be for the exact same price while they could still make a huge margin. They are purposefully holding us back, it could be better, much better.

Your argument is we should be happy because it could be worse? I say we should be angry because it could be better.

Comment Re:Yeah, automated tweeting to PR mouthpiece... (Score 1) 153

Mine didn't. Not only do they not use the term "up to", but they explicitly point that out and claim you will always get 100% of your bandwidth *within their network and to their peers and trunk. $45/m for 100/100 and currently 14ms from Chicago. Used to be 6ms until Level 3 changed routes. uhggg. fml

Comment Re:RAID 0 is not for anything you don't want to lo (Score 1) 73

Correct. SSDs fail for different reasons than spinning rust. Most mechanical HDs fail for physical reasons and physical reasons tend to be highly correlated for all drives in an array, even if they're different models or even brands. There is a very high risk that if one drive fails, another is right behind it. RAID5, I'm looking at you.

RAID0 is any drive failing is a loss, so multi-drive failures don't matter so much, but they're also much less likely until it's a firmware bug or other pathological issue. But SSDs are pretty much RAID0 already and have a fraction the failure rate of mechanical drives.

Comment Re:This is retarded. (Score 1) 73

M.2 supports up to 2^16 queues with a queue depth of 2^16 each. I wonder when SSD controllers will start to take advantage of this new protocol. These first gen M.2 drives seem more like retro-fitted SATA drives that just so happen to have beefy controllers, but were not specifically designed to make full use of the new features.

Comment Re:Oh boy! (Score 4, Interesting) 204

Many Youtube videos have already been taken down because they have mentioned the words "react" or "reaction" over this. They're also quickly tracking down Youtube videos that use the base word "react" in any way and asking for a cut of ad revenue or be taken down. It's not only an overreaching trademark, but it's being quickly abused.

Comment Re:Isn't this what --preserve-root is for? (Score 1) 698

POSIX may have changed, but at one point, running the idea of not allowing "rm -rf /" to work by default was heresy to an official POSIX expert until someone realized that part of the standard said if /dev (or something, I forget what path), would be a target of the pattern, then the entire call is undefined as per the standard. Solaris eventually added a protection to refuse to run, but different *nix systems handle it differently.

Slashdot Top Deals

The person who's taking you to lunch has no intention of paying.