Qualcomm Unveils 24 Core ARM Server Processor

Ever since ARM took over the mobile world, we’ve been hearing about how the RISC architecture was trying to expand into other higher performance areas. While ARM first entered the server world via relatively weaker micro-servers, it looks like performance options are coming along soon. Qualcomm, one of the major ARM partners and licensees, has unveiled their first 24-core server processor.

Running off the latest ARMv8-A instruction set, the chip will be part of Qualcomm’s Server Development Platform and part of a stack of server tools. While details on the chip itself have been slim, we do know that it is based off a FinFET process, meaning either TSMC 16nm or Samsung 14nm. The core is also fully custom meaning it is not like the A57/A53 found int he Snapdragon 808/810 and given the server environment, probably isn’t Qualcomm’s in-house 64bit Kyro architecture found in the Snapdragon 820.

Along with the Soc, the SDP  also includes server-class PCIe and storage interconnects. Other hardware requirements like ethernet and FPGA are to be proved by Mellanox and Xilinix respectively. On the software side, SDP comes with a software stack capable of running a KVM Linux hypervisor, OpenStack DevStack, and guest Linux distributions running Apache and WordPress as shown in the demo.

With a complete package available, Qualcomm stands a good chance at breaking into the server market. The biggest question is if ARM will be able to bring the same performance to the table as x86 does while still remaining more power efficient and cheap. This is especially true of single threaded IPC where Intel has historically dominated.

GIGABYTE Announces 384 Cores Server Powered by Cavium ThunderX ARMv8 Processors

GIGABYTE doesn’t just build consumer products, they also got a nice line-up of servers and motherboards for server usage. They have just announced the new server board MT70-HD0 and associated 2U rackmount server H270-T70 and this is the first cache-coherent dual socket ARM motherboard to be released to the market.

The new model is built around the Cavium ThunderX ARMv8 processor with 48 cores and running at 2.5GHz. With two processors per board and four nodes per server, we reach an astonishing 384 total cores in a single 2U rack unit. That’s quite impressive.

Memory shouldn’t be an issue for all these cores as each of the nodes supports up to 1TB high-speed 2400MHz DDR4 memory. The two 40Gb/s LAN ports are directly controlled by the SoC processors and the unit supports full IPMI based remote board management.

The H270-T70 comes with two redundant 1600W PSUs with 80 Plus Platinum certification. There are also plenty of USB 3.0 ports as well as front and rear VGA ports for maintenance jobs.

Low SoC TDP combined with high end onboard memory and networking enables a platform with enormous potential for operators of core-intensive applications looking for uncompromised performance and significant power savings simultaneously. The MT70-HD0 and the H270-T70 will be both available for volume shipment in the third quarter of 2015.

AMD’s Official Roadmaps Reveals the Company’s Plans for the next 5 Years

AMD has revealed what the company plans to do with its GPUs and CPUs in the next 5 years at the PC Cluster Consortium event in Osaka Japan, where AMD’s Junji Hayashi revealed the company’s roadmap.

During the event, AMD has focused on its graphics IP and the products that involved it, including discrete Radeon graphics cards and Radeon powered Accelerated Processing Units. There have been talks about AMD’s upcoming K12 ARM as well as the x86 Zen CPU core, including a strategy of how the company plans to introduce both x86 and ARM powered SOCs to the market in a pin for pin compatible platform code-named SkyBridge.

It is said that both CPUs are 64-bit capable parts coming in a 14nm FinFET ‘shell’, but one is based on the ARMv8 architecture while the other is based on the more traditional x86 AMD64 architecture, having them target the server, embedded, semi-custom and client markets.

AMD has also talked about “many threads” revealing that the K12 will come with Simultaneous Multi-Threading (SMT) technology in contrast to the company’s Clustered Multi-Thread (CMT) technology we are able to see in the Bulldozer family. SMT essentially takes advantage of the various resources in the core which are underutilized and dedicate to an additional, slower, execution thread for added throughput. In contrast, CMT is looking for opportunities to share resources between two different CPU cores, instead of doing it inside a single CPU core.

Hayashi also revealed AMD’s GPU roadmap, which reveals that the company is employing a two-year cadence to updating its GPU architecture inside APUs. It looks like the company will add Accelerated Processing Units with updated GPU architectures once every two years. The roadmap also reveals that AMD plans to introduce what it described as a High Performance Computing APU which carries a 200 – 300 watts TDP, having the company stating that the APU in question will excel in HPC applications.

AMD apparently did not attempt to go with newer APUs in the future because it was not viable in terms of memory bandwidth. Instead, the company’s stacked High Bandwidth Memory will be used as an alternative, making the design extremely effective. The second generation of HBM is said to be 9 times faster than GDDR5 memory and 128 times faster than DDR3 memory.

The company has not revealed any code names for the GPU architectures, but a previous leak pointed out that the architecture will debut on 16nm FinFET and will be code-named Arctic Islands. More specific details about AMD’s products will be revealed in May at the Financial Analyst Day event.

Thank you WCCF for providing us with this information

Latest Gigabyte microATX Motherboard Arrives with Octo-Core ARMv8 SoC

Gigabyte apparently has developed a new microATX motherboard that comes with a 64-bit ARM onboard processor, a motherboard that is dedicated to datacenters.

The MP30-AR0 board is said to be built around Applied Micro’s X-Gene SoC, which is an octo-core chip clocked at 2.4 GHz that comes with a 45W TDP. Applied Micro’s cores are said to be their own design and are compatible with the 64-bit ARMv8 ISA.

The SoC appears to be paired to a quad-channel memory controller with eight UDIMM slots, each of them supporting modules of up to 16GB at 1600 MHz speeds and is also ECC compliant. The board is not designed with a great graphical solution in mind, having the board be more networking oriented.

The motherboard is said to come with dual 10-Gigabit Ethernet controllers embedded in the SoC and two auxiliary Marvell Gigabit Ethernets on the board itself. In terms of storage solution, the MP30-AR0 comes with only four SATA 6 Gbps ports and one SD slot. Two PCIe x16 slots are also present on the board, but each slot provides eight lanes of Gen3 bandwidth, most likely from the SoC.

Gigabyte states that the motherboard supports Ubuntu 14.04, having the motherboard also available in the company’s R120-P30 1U server.

Thank you TechReport for providing us with this information

Qualcomm Snapdragon 610 and 615 Announced for Q3 2014, 801 Expected This Quarter

Qualcomm has revealed three new SoCs during MWC, the new Snapdragon 610 and Snapdragon 615, both featuring the ARMv8 64-bit 28nm processing technology and the third generation LTE modem for 4G connectivity and dual-SIM support. The last, but not least, Snapdragon revealed is the high-end Snapdragon 801 having several enhancements compared to its Snapdragon 800 predecessor.

“One thing that’s going to be a hallmark of our Snapdragon story is that we’re going to be providing Snapdragon solutions that essentially maximise our user experience at every tier so smartphone can prevail,” said Qualcomm EVP Murthy Renduchintala at MWC in Barcelona today.

The main difference between the Snapdragon 610 and Snapdragon 615 is the number of cores, having the 610 coming in with dual-core technology and 615 providing quad-core technology. Besides that, both Snapdragon SoCs feature the same Andreno 405 GPU found on the Snapdragon 800, having support for DirectX 11.2 and Open GL ES 3.0, with added support for hardware accelerated geometry shading and hardware tessellation, which Qualcomm said will provide more detailed, realistic mobile games and user interfaces.

The Snapdragon 801 features higher quality imaging with support for larger, faster camera sensors and improved image post-processing, as well as higher speed SD card memory support and dedicated dual SIM hardware support. As the Snapdragon 800, the Snapdragon 801 includes 4G LTE and 802.11ac Wi-Fi connectivity , in addition to a quad-core Krait 400 processor for advanced performance and an Adreno 330 GPU for premium graphics.

“The Snapdragon 801 is a key platform for us, key drivers are to improve user experiences, [include support for] h265 video, supporting more sophisticated camera solutions and dual SIM configurations,” said Renduchintala. “The camera performance is, from a sensor point of view, 45 percent faster, the graphics solutions have been improved by over 28 percent, [with a] 14 percent CPU capability improvement as well as 14 faster memory solutions.” he added.

While the Snapdragon 610 and 615 processors are expected to be released in the third quarter of 2014, the Snapdragon 801 is expected to be available in consumer handsets starting this quarter.

Thank you The Inquirer for providing us with this information

Nvidia’s 64-bit Tegra 6 To Power The First 64-bit Android Devices

Thanks to a report from ExtremeTech, NVIDIA may have bumped up their release roadmap by sucking Tegra 6 aka “Parker” from 2015. Instead, Parker will see the light of day in late 2014 itself, and will also be the first 64-bit NVIDIA chipset. The report is based on new findings which indicate NVIDIA licensed and began working on ARMv8 64-bit architecture back in late 2010/early 2011 itself. 64-bit “Denver” cores should be ready for sampling in early 2014 itself.

No one can deny that Tegra 4 is a much more competitive chip than the Tegra 3 was last year, but Qualcomm’s custom Krait cores and quick design iteration have allowed it to nearly take over the Android device ecosystem. There are only a handful of devices announced with Tegra 4 chips, and one of them is Nvidia’s own Shield console. Meanwhile Snapdragon 600 and 800 power almost every high-end smartphone and tablet from 2013.

Over the last few years, Nvidia has no doubt poured many millions of dollars into Project Denver and the Parker SoC. Qualcomm is surely working toward a 64-bit chip down the road, but the company has been tight-lipped about future plans. The last thing Nvidia wants is for the market leader to beat it to ARMv8 with 64-bit. Whether or not 64-bit has real utility on mobile devices running 32-bit software, getting this right could set up Tegra for undeniable success, a distinction that has so far eluded it.

Thank you NextPowerUp and ExtremeTech for providing us with this information.

Images courtesy of ExtremeTech.