AMD made an interesting enterprise announcement today with the introduction of new FirePro S-Series graphics cards that integrate hardware-based virtualization technology. The new FirePro S1750 and S1750 x2 are aimed at virtualized workstations, render farms, and cloud gaming platforms where each virtual machine has direct access to the graphics hardware.
The new graphics cards use a GCN-based Tonga GPU with 2,048 stream processors paired with 8GB of ECC GDDR5 memory on the single slot FirePro S1750. The dual slot FirePro S1750 x2, as the name suggests, is a dual GPU card that features a total of 4,096 shaders (2,048 per GPU) and 16 GB of ECC GDDR5 (8 GB per GPU). The S1750 has a TDP of 150W while the dual-GPU S1750 x2 variant is rated at 265W and either can be passively cooled.
Where the graphics cards get niche is the inclusion of what AMD calls MxGPU (Multi-User GPU) technology which is derived from the SR-IOV (Single Root Input/Output Virtualization) PCI-Express standard. According to AMD, the new FirePro S-Series allows virtual machines direct access to the full range of GPU hardware (shaders, memory, ect.) and OpenCL 2.0 support on the software side. The S1750 supports up to 16 simultaneous users and the S1750 x2 tops out at 32 users. Each virtual machine is allocated an equal slice of the GPU, and as you add virtual machines the equal slices get smaller. AMD’s solution to that predicament is to add more GPUs to spread out the users and allocate each VM more hardware horsepower. It is worth noting that AMD has elected not to charge companies any per-user licensing fees for all these VMs the hardware supports which should make these cards more competitive.
The graphics cards use ECC memory to correct errors when dealing with very large numbers and calculations and every VM is reportedly protected and isolated such that one VM can not access any data of a different VM stored in graphics memory.
I am interested to see how these stack up compared to NVIDIA’s GRID and VGX GPU virtualization specialized graphics cards. The difference between the software versus hardware-based virtualization may not make much difference, but AMD’s approach may be every so slightly more efficient with the removal of layer between the virtual machine and hardware. We’ll have to wait and see, however.
Enterprise users will be able to pick up the new cards installed in systems from server manufacturers sometime in the first half of 2016. Pricing for the cards themselves appears to be $2,399 for the single GPU S1750 and $3,999 for the dual GPU S1750 x2.
Needless to say, this is all a bit more advanced (and expensive!) than the somewhat finicky 3D acceleration option desktop users can turn on in VMWare and VirtualBox! Are you experimenting with remote workstations and virtual machines for thin clients that can utilize GPU muscle? Does AMD’s MxGPU approach seem promising?
You can buy 10 r380s for the
You can buy 10 r380s for the price of one s1750
But only ten users can be
But only ten users can be fed.
[sidenote]
I’m sorry, I meant
[sidenote]
I’m sorry, I meant you can buy 10 R380Xs which have also 2048 stream processors.
[/sidenote]
Yes, but these 10 users will get 2048 stream processors each. But I understand your point, virtualization has benefits and therefore is worth more, but 10x more?
I do agree the prices are a
I do agree the prices are a bit high considering they are based on tonga. If they were based on Polaris with HBM 2 memory then i could understand
Once again for the billionth
Once again for the billionth time its the ECC memory and other PROFESSIONAL features like graphics drivers certified to work with the PROFESSIONAL graphics software that makes these cards costly! So they are not for gaming they are for professional/scientific use, where errors can not be allowed. Even the workstation’s mainboard will support ECC memory for the CPU part of the system in addition to the GPU card’s ECC Graphics memory! The costs of professional graphics software certification does not come cheap! So once again gamers Go Figure! But in large part it’s the software/driver certification that costs a lot and makes Pro graphics cards costly, in addition to all the ECC memory ability!
Removing that extra software
Removing that extra software later is worth GOLD for Workstations, is huge news! Need to divide the cost by 16 or 32. Just in tech super salary saved it pays for itself with a year plus ease of maintenance and resource allocation. AMD looks really good here!
That is awesome. One gaming
That is awesome. One gaming machine to feed a whole house full of low-power gaming terminals. Can’t wait for something like that to trickle down to consumer cards.
It’s not for gaming, it’s
It’s not for gaming, it’s more for precision and accuracy, it’s not quick and dirty gaming graphics where a few bits dropped on a 60+ frame per second render rate will not be noticed by any gamer. It’s for those multiple hour renders of a single image that my be blown up to outdoor advertising size, or large trade show booth graphics size where a few dropped pixels will be very noticeable! Same goes for scientific and engineering work where a simple error could mean that the aircraft’s rudder may fracture or other important part may stop working and lead to disaster. You would not want your CPU/GPU closed coolant pump, and pump housing designed with a gaming GPU, do you even know how much the mathematical libraries/drivers are gimped for Gaming calculations to get those quick and dirty graphics calculations done to render a game at 60+ FPS, you would not want any engineering work done using that software, not to mention the lack of ECC memory ability to prevent even more erroneous results!
P.S. Imagination Technologies uses: “virtualization technology to create up to 8 zones that are fully isolated from each other.”(1)
(1)
http://www.anandtech.com/show/8706/imagination-announces-powervr-series7-gpus-series7xt-series7xe
Doesn’t matter what it’s
Doesn’t matter what it’s designed to do, I would use it for gaming.
Maybe the GPU virtualization
Maybe the GPU virtualization technology, but not the professional math libraries and certified drivers, gaming has no need for that much mathematical accuracy, and error free rendering! Gaming only needs speed. So maybe having a home based server that can run more than one game at a time to multiple users via GPU virtualization technology. Just be sure to have enough GPU capacity in the server if more than one user needs to play a AAA title on ultra settings at say more than 1080p!
These are obviously NOT
These are obviously NOT gaming cards.
/facepalm
Hey Tim, just saying that
Hey Tim, just saying that there’s a typo in this post; where it says “2,098 shaders”, this should be 2,048.
Yup, you’re right, it’s 2048.
Yup, you're right, it's 2048. Thanks for pointing that out!