eMAG is a family of high-performance ARM server processors designed by Ampere Computing. Ampere’s introduction of eMAG to the market concludes and follows on the X-Gene3 design started out by AppliedMicro. eMAG processors targets server workloads capable of taking advantage of a high core count with high throughput. First generation eMAG processors are based on the Skylark microarchitecture, a design that started out by AppliedMicro. Fabricated on TSMC’s 16FF+ process, those processors feature up to 32 cores operating at up to 3.3 GHz. DDR4 channels, up to 2666 MT/s with ECC; 1 TiB/socket I/O: 42 PCIe Gen 3 lanes TDP: Up to 125 W Second generation eMAG processors are planned for 2019. Those chips will be based on Ampere’s Quicksilver microarchitecture and feature an array of new features and improvements developed also with the new staff that Ampere hired over from Qualcomm’s ARM Server team.
Jon Masters, Chief ARM Architect at RedHat and Yan Fisher, RedHat Hyperscale Platforms Product Marketing Lead, talk about the traction in the industry for the ARM Servers, where Red Hat talks about running the same unmodified OS on 10 different ARM SoC vendor’s hardware, where Red Hat expects significant adoption in the next 12 months as customers ask for Linux on ARM, Red Hat can help these customers deploy, support and upgrade. As next generation better ARM Server solutions are getting ready to be deployed. ARM was talking about 20% of new servers by 2020 could be ARM Servers, Red Hat is bullish and thinks it could be higher.
OPNFV Pharos Lab project deals with developing an OPNFV lab infrastructure that is geographically and technically diverse. The Pharos Lab is hosted in Kista, Sweden, it will greatly assist in developing a highly robust and stable OPNFV platform (see more: https://wiki.opnfv.org/pharos) OPNFV is a carrier-grade, integrated, open source platform to accelerate the introduction of new NFV products and services (see more: https://wiki.opnfv.org/start)
The following ARMv8 servers are used:
– Controller nodes: 3 * Applied Micro X-Gene 2 ARMv8-64 8 cores @ 2.4GHz, 32GB RAM, 1x128GB SSD, 2x1TB HDD, 1x10Gbps SFP+ NICs, 2x1Gbps NICs.
– Compute nodes: 2-3* Cavium Networks CN8890-CRB ThunderX ARMv8-64 48 cores @ 2.5GHz, 8x16GB RAM (128GB total), 1x500GB HDD, 1x40Gbps QSFP+ NIC, 2x10Gbps SFP+ NICs, 1x1Gpbs NIC (RJ45, IPMI interface).
ENEA’s demo in ARM booth was showing a simple NFV application running on our operational ARMv8 Pharos lab infrastructure. The application demonstrates a simple NFV service chain integrating a DPI (deep packet inspection) VNF engine provided by QOSMOS (see more: http://www.qosmos.com).
Gigabyte launched their AppliedMicro X-Gene 1 server on 64bit. This server can connect into for example the Annapurna storage server. By October Gigabyte will also be shipping the AppliedMicro X-Gene 2 server with DDR4 speed, lower power consumption, where AppliedMicro designs their custom SoC with many features integrated and performance and power consumption optimized. You can also see my tour at AppliedMicro featuring the X-Gene 1 and X-Gene 2: http://22.214.171.124/2014/12/11/how-the-appliedmicro-x-gene-arm-server-processors-are-designed/
Gigabyte is launching a whole range of ARM Powered Servers at Computex 2015: Gigabyte H279-T70 based on the Cavium ThunderX with 384 cores in a 2U system, Gigabyte D120-S3G featuring the Annapurna Labs Alpine AL5140 quad-core ARM Cortex-A15 and the Gigabyte R120-P30 is based on the Applied Micro X-Gene 1 Octa-core 64bit processor.
In this video, AppliedMicro’s Kumar Sankaran discusses the software of the X-Gene platform and provides a comparison of X-Gene 1 and 2 against the latest Intel server processors Xeon E5.
You can order AppliedMicro’s 64bit ARMv8 development board here: https://www.apm.com/products/data-center/x-gene-family/x-c1-development-kits/
AppliedMicro gives an overview of X-Gene, providing the different design components as well the various benefits in using X-Gene for compute server, storage and high performance computing. AppliedMicro is one of the initial partners with ARM in developing the ARMv8 64bit architecture, and customizing it for high performance server computing.
AppliedMicro’s Gaurav Singh gives us a sneak peek into the development labs of X-Gene 2 showing a live demonstration ready for production, with AppliedMicro X-Gene 2 coming out for ARM Servers in 2015.
World’s first 64bit ARMv8 development board (you can order it here: https://www.apm.com/products/data-center/x-gene-family/x-c1-development-kits/) based on the Octa Core X-Gene 2.4Ghz running in SMP mode available for anyone to buy today. It’s built for Servers, supports 64bit Android development, featured in the HP Moonshot ARM Server product. Designed for cloud computing and next-generation data centers, featuring custom high-performance ARMv8 cores, AppliedMicro X-Gene is the first to couple an advanced 64-bit ARM architecture with unique network and storage offload engines, as well as integrated Ethernet. The highly integrated, purpose-built X-Gene solution delivers the highest performance and lowest total cost of ownership (TCO) for private cloud, public cloud, and enterprise applications.
Clark and Linda of HP give an inside look at HP’s Moonshot system configured with their new m400 ARM cartridges. Each cartridge is an individual 64-bit ARM server using AppliedMicro’s X-Gene SOC, with 8 cores and 64Gb of RAM with 2 Mellanox 10G NICs. The servers are running OpenStack with a mix of cloud controller services and Nova compute nodes.
Linda from HP describes HP’s new Moonshot systems, including the new m400 ARM server cartridge, which was demoed at Linaro Connect. HP has launched the TI 32bit and the AppliedMicro X-Gene 64bit ARM Server in HP Moonshot.
Jon Masters, Chief ARM Architect at Red Hat, talks about Red Hat showing off their ARM Partner Early Access Program running on AMD’s ARMv8 64bit Seattle and on the Applied Micro ARMv8 64bit X-Gene Mustang booting both with UEFI and ACPI on a single same Kernel with no changes, common platform. Jon Masters talks about the Linaro Enterprise Group’s status and how much is yet required to be done for ARMv8 Servers to get into mass deployments worldwide.
Linaro’s validation team demonstrates using LAVA for native toolchain builds and “hack” sessions on Arndale Octa and APM X-Gene ARMv8 platforms
Linaro’s automated validation architecture (LAVA) is typically used to execute automated tests to validate Linaro’s engineering output. However, LAVA has recently integrated features to automate builds and provide secure remote interactive sessions to developers. Linaro’s lab lead Dave Pigott shows a native toolchain build orchestrated by LAVA. This technology enables developers to validate the toolchain on many ARM processor designs as well as other architectures. Tyler Baker a technical architect at Linaro explains how LAVA abstracts the image deployment, boot process, and installation of software needed to support these “hack” sessions.
Showing off the latest Applied Micro 64bit X-Gene ARM Server Development Board (which Rob Savoye of Linaro eagerly wants to start playing with), Dell’s 64bit ARM Server solution running Fedora 19, working on a proof of concept for early 2014 for Dell’s key cloud server customers (Google? Amazon?) before going into mass production (Dell already did some 32bit ARM Server tests in Europe with some customers), some things like Oracle JDK still has to fully come over (needs some tuning) to the platform. This is just me walking kind of randomly around some of the ARM Server demo area at the ARM Techcon. Then checking out the HP ARM Server booth, showing off some of the latest HP Moonshot ARM Server solutions also talked about in HP’s Keynote at ARM Techcon, watch the official video of that keynote or my version (sitting on the front row).
Red Hat is working on getting Fedora Linux up and running on ARMv8, Jon Masters talks about LEG Linaro Enterprise Group and getting the full Fedora Linux stack to work on 64bit ARMv8 server hardware from now and during next year as 64bit ARMv8 hardware becomes available. AMD, AppliedMicro, Calxeda, Canonical, Cavium, Facebook, HP, Marvell and Red Hat join existing Linaro members ARM, HiSilicon, Samsung and ST-Ericsson to form new group focused on accelerating Linux development for ARM servers.
Vinay Ravuri, Vice President and General Manager, Server Products at AppliedMicro gives an update on the 64bit ARM X-Gene Server Platform. At ARM Techcon 2012, AppliedMicro, ARM and several open-source software providers gave updates on their support of the ARM 64-bit X-Gene Server on a Chip Platform.
At ARM TechCon 2011 last week, Applied Micro was able to show their ARMv8 platform design already running on an FPGA, to be sent out to their partners in January so they can start working on the software for when they can have working silicon of their ARMv8 64-bit Server-on-chip platform, they say as early as in the 2nd half of 2012 already. Here is the full keynote presentation featuring Paramesh Gopi, president and CEO of Applied Micro, Lance Howarth, EVP Marketing at ARM, Dr. Christos Kozyrakis of Stanford University, Andrew Feldman, Founder and CEO of SeaMicro and Vinay Ravuri, Vice President of AppliedMicro’s Embedded and Processing Business Unit, presenting the worlds first ARMv8 64-bit processor demo running on an FPGA. I recommend that you watch the full webcast with slides on Applied Micro’s own website (enter a name and email to start watching in full screen with the synchronized slides), and here is the YouTube version without the slides as published by youtube.com/cnxlinux:
One can thus possibly understand from this that the ARM Powered Servers are going to be upgraded twice in the next year. Powered by Quad-core ARM Cortex-A9 now such as the HP Moonshot project powered by Calxeda EnergyCore, likely upgraded to ARM Cortex-A15 solutions (up to 8 cores) as soon as those are ready (2H 2012) and then again upgraded to ARMv8 64-bit running at up to 3Ghz which is what Applied Micro is saying that they can deliver early silicon of in just about a year from now. Thus ARM Powered Servers are going to run at up to full performance levels, not only being suitable for lower power consumption and lower price but also aiming to deliver the full maximum performance that some people building servers say they need.
- Applied Micro leaps ahead in ARM server race (go.theregister.com)
- 64-bit ARMv8 architecture to debut in AppliedMicro X Gene SoC (linuxfordevices.com)
- ARM targets 64-bit servers (h-online.com)
- Friday Video (late): Want more info on the AppliedMicro 64-bit ARM v8 X-Gene server SoC? Photos and a link to a video (eda360insider.wordpress.com)
- AppliedMicro demos FPGA emulation of multicore server chip based on new 64-bit ARMv8 architecture (eda360insider.wordpress.com)
- ARMv8 detailed: 64-bit architecture, AppliedMicro first in line (engadget.com)