8 ms·
Microsoft Readies Azure GPUs
- mrmondo 10y agoMeanwhile Azure and Office 365 still use Symantec issued SHA1 certificates which are insecure, Symantec owns bluecoat who sell interception technologies to governments and private enterprises, all to convenient given that Microsoft themselves are a CA and could have issued their own certs, handing that off to a company like Symantec means they aren't directly giving your data away and are avoiding direct blame as much as possible.
- dogma1138 10y agoSSL Certificates are a legal tool these days, Microsoft doesn't want to deal with it, the insurance and validation is handled by Verisign/Symantec it's a business decision. Bluecoat is hardly an "interception" technology it's a mid 90's enterprise grade proxy and internet policy management box that can do SSL interception their early boxes used SSL strip the newer ones probably still do. They don't have a magic backdoor to break SSL all you can do with a Bluecoat box is to install your own enterprise certificate and make it act as an SSL proxy to ensure that your employees do not surf pronhub over SSL at work. So as far as tinfoil hat rants goes i give this one 3/10 it's not even that creative, you might want to look the "PRISM" partners slides again might give you some inspiration.
- exhilaration 10y agoUpvoted for the timely Olympic scoring.
- Artemis2 10y agoIsn't Microsoft's CA used only to sign their software?
- n00b101 10y agoNot sure about the exact pricing shown here, but it looks like Azure blows AWS GPU pricing out of the water. Comparing specs: AWS g2.2xlarge: * DRAM: 15GB * GPU processor: 1xGK104 * 1536 CUDA cores * 2.29Tflops/s peak single precision throughput * GPU memory: 4GB * 160GB/s peak bandwidth * Price: $0.65/hr (US East, Linux) Azure NC6: * DRAM: 56GB * GPU processors: 2xGK210 * 4992 CUDA cores * 8.73Tflops/s peak single precision throughput * GPU memory: 24 GB * 480GB/s peak bandwidth * Price: $0.66/hr I expect AWS will introduce a new GPU fleet and reduce pricing of the existing G2 fleet, in response to Azure.
- JoshGlazebrook 10y agoI wish one of the main cloud platforms would significantly cut their per/GB bandwidth costs. It feels like there has been no adjustments in years. Also funny story, I was 14 or 15 and was trying to figure out a name for a stupid little chat client I was trying to make in C#, and somehow azure popped into my head. Not surprisingly azure.com was taken and just had a generic microsoft landing page (this was a little before it was announced in 2008).
- blahi 10y agoThat's a major lock-in mechanism. Data stays in for the compute. Which is a major pity for Microsoft since all the marketing measurement startups got built on top of AWS while Azure has superior analytics tools.
- snuxoll 10y agoI mean, it doesn't necessarily solve anything if you're using it for public facing apps, but if you're doing a hybrid solution with Azure you can buy unmetered ExpressRoute ports and at least avoid the metered bandwidth costs back to a datacenter.
- dharma1 10y agoit sure does. I hope AWS and GCE will step up their GPU instance game
- dharma1 10y agoI signed up for the beta but haven't heard anything back yet. Anyone else?
- gwern 10y agoI haven't either.
- wjoe 10y agoThis is pretty interesting from a video encoding point of view. We're currently using g2.2xlarge AWS instances which are Kepler based. All we really need from those instances is the GPU (more specifically, the NVENC H264 encoding chip), so other specs aren't too important. It looks like the NV6 Azure instances use a newer Maxwell based chip, which can handle much more (concurrent, or higher bitrate/resolution) videos, compared to the 2 generation older AWS GPUs. And the NV6 is virtually the same price as the g2.2xlarge... time to do some testing. Hopefully Amazon can catch up soon.
- dogma1138 10y agoJust for reference: K1 (AWS): 1 GPUs x 6 H.264 streams (max single stream 720p@30fps) K2 (Azure K80): 1 GPUs x 6 H.264 streams max single stream 720p@30fps) M6 (Don't know if it's used by anyone ;)): 1 GPU x 18 H.264 streams (max single stream 1080p@30fps) M60 (Azure M60): 1 GPU x 18 H.264 streams (max single stream 1080p@30fps) You can sacrifice encoder bandwidth (max streams) for higher resolutions or frame rates.
- zip1234 10y agoIs GPU better than CPU for encoding? I thought performance ended up similar. Also, are you saying that K1 AWS can reencode 6 streams at once?
- dogma1138 10y agoThe performance of the GPU is orders of magnitude better you can do multiple 1080p streams in real time with high bandwidth easily. The quality on the other hand even for the same bit rate is considerably worse than highly optimized CPU compilers as the GPU does do a lot of hacks along the way.
- valarauca1 10y agoGPU generally support more codec and faster encoding/decoding times then a CPU. But it really depends on what codec, what CPU, and what GPU.
- Athas 10y agoThis is interesting. Is there some way for academics to get discounted access for experiments (not for running heavy simulations, but for testing whether our stuff works)?
- dogma1138 10y agoMicrosoft has graned "free" subscriptions to proffessors and researchers before, so it can't hurt to ask :P https://azure.microsoft.com/en-gb/community/education/ https://azure.microsoft.com/en-gb/community/education/
- LogicFailsMe 10y agoAWS: 4 year-old GPUs Azure: 2 year-old GPUs Wouldn't it be awesome if a cloud provider could keep up with the GPU roadmap and ship something within a year or so of its release?
- deelowe 10y agoI think supply is the issue... So nvidia and amd are the problem.
- perryh2 10y agoThe GPUs that hosting providers are buying need to support vGPU features, which are only support by the higher-end workstation and server products.
- LogicFailsMe 10y agoWhich is a software/binning issue (probably 100% software) not inherently HW because M60==GTX 980==GM204 and M40==GTX Titan X==Quadro M6000==GM200. Interestingly, for the first time ever, GP100 is unique (well, OK, K80 too but K80 was too late). And since Quadro P6000==Titan XP==GP102, it's probably just a software block here as well. Also, for the first time ever, the high-end Quadro will be the best FP32 GPU of it's generation. That's the most interesting part for me.
- dogma1138 10y agoIt is a software/bios block. Once the driver grabs a desktop card and it's initiated it's locked. The cards also come with a UEFI bios which means if your host initialized the card during boot it's also locked. To overcome this you need to disable UEFI boot/GPU boot in the BIOS and blacklist the card in the host OS and then create a PCI-stub device that will be used for the passthrough. This is an utter and complete hack and you can't really use for production grade implementations. I'm not sure that Quadro actually supports vGPU also AFAIK only Tesla and Grid parts do, Tesla does have some additional in GPU support for virtualization, i don't know how GRID handles it. GRID parts are less for compute and more for game/video streaming so they aren't fully virtualized, AFAIK they do not support P2P gpu communication or shared memory access, they are only more Hypervisor friendly so they can be thin provisioned.
- gwern 10y ago/r/machinelearning discussion: https://www.reddit.com/r/MachineLearning/comments/4wd5kn/ms_expands_cloud_gpu_offerings_k80s_at_056hour/ https://www.reddit.com/r/MachineLearning/comments/4wd5kn/ms_...
- Dolores12 10y agoAny idea if it worth to run a bitcoin miner on those like it used to be few years back?
- kim0 10y agoDoes anyone know of an access protocol (like pcoip) that supports multiple users to have concurrent 3d accelerated sessions on one VM ? Each just 720p