Posted on

Arm Server Update, Spring/Summer 2021

As usual, we are overdue for an update on all things Arm Servers! Today’s announcement of the Arm v9 specification is a great time to review the state of Arm Servers, and what has changed since our last update.

First, let’s review our last update. Marvell canceled the ThunderX3 product, Ampere had announced the Altra but it wasn’t shipping, AWS Graviton was available, and Nuvia was designing a processor.

Fast forward to today, and the Ampere Altra’s are now becoming available, with limited stock via the Works on Arm program at Equinix Metal, and some designs shown off by Avantek, a channel supplier. Mt. Snow and Mt. Jade, as they are known, are also formally designated as “ServerReady” parts, passing standards compliance tests.

Nuvia, the startup that was designing a new Arm Server SoC from the ground up, was purchased by Qualcomm, in an apparent re-entry into the Arm Server market (or for use in Windows on Arm laptops?). Don’t forget, they previously had an Arm Server part, the Centriq, though they scrapped it a few years ago. So, it now remains to be seen if Nuvia will launch a server-grade SoC, or pivot to a smaller target-device.

The other emerging trend to cover is the role of Arm in the Edge Server ecosystem, where the trend of pushing small servers out of the datacenter and closer to customers and users is rapidly gaining momentum. In this scenario, non-traditional, smaller devices take on the role of a server, and the energy efficiency, small form-factor, and varied capabilities of Arm-powered single board computers are taking on workloads previously handled by typical 1U and 2U rackmount servers in a datacenter. But, small devices like the Nvidia Jetson AGX, RaspberryPi Compute Module 4, and NXP Freeway boxes are able to perform Edge AI, data caching, or local workloads, and only send what is necessary up to the cloud. This trend has been accelerating over the past 12 – 18 months, so, we may see some more niche devices or SoC’s start to fill this market.

Posted on

Arm Server Update, Fall 2020

The announcement yesterday of the cancelation of Marvel’s ThunderX3 Arm Server processor was a reminder that we were overdue for an Arm Server update!  So, continuing on in our regular series, here is the latest news in the Arm Server ecosystem.

As mentioned, unfortunately it appears at this time that Marvell has canceled the ThunderX3 Arm Server processor that was shown earlier this year, and would have been the successor to the ThunderX and ThunderX2 parts released previously.  The current rumors indicate that perhaps some specialized version of the SoC may survive and be used for an exclusive contract with a hyperscaler, but that means “regular” customers will not be able to acquire the part.  And with no general purpose, general availability part, the ThunderX3 will effectively be unavailable. 

That leaves AWS providing the Graviton processor in the EC2 cloud server option, or Ampere with their current generation eMag Arm Server, and forthcoming Ampere Altra SoC as the only server-class Arm processors left (for now).  The Ampere Altra is brand new, and available from our friends at Packet in an Early Access Program, but no specific General Availability date has been mentioned quite yet.  This processor offers 80-cores or 128-cores, and is based on Arm Neoverse N1 cores. 

There is another processor on the horizon though from Nuvia, a startup formed late last year who is designing an Arm-based server class SoC.  Nuvia has said it will take several years to bring their processor to market, which is a typical timeframe for an all-new custom processor design.  So in the meantime, only Amazon and Ampere are left in the market.

The NXP desktop-class LS2160 as found in the SolidRun Honeycomb could also be considered for some workloads, but it is a 16-core part based on A72 cores.

There is one other Arm Server that exists, but unfortunately it’s not able to be acquired outside of China:  the Huawei TaiShan 2280 based on the HiSilicon Kunpeng 920.  This is a datacenter part that is likely used by the large cloud providers in China, but seems difficult (or impossible) to obtain otherwise.  It is a dual processor server, with 64-cores in each processor, thus totaling 128 cores per server.

As usual, the Arm Server ecosystem moves quickly, and we look forward to seeing what’s new and exciting in our next update!

 

Posted on

Where to Buy an Arm Server

Being Arm enthusiasts and deeply embedded in the Arm Server ecosystem, one of the questions we get asked often is,

“Where can I buy an Arm Server?”

In the past, it was difficult to actually find Arm Server hardware available to individual end-users. Not long ago, the only way to gain access to Arm Servers was to have NDA’s with major OEM’s or having the right connections to get engineering-sample hardware. However, over the course of the past 2 to 3 years, more providers have entered the market and hardware is now readily available to end users and customers. Here are some of the easiest ways to buy an Arm Server, although this list is not exhaustive. These servers all have great performance, relatively low costs, and are well supported thanks to standards compliance and UEFI.

First and foremost, the AMD Opteron A1100 may not be a commercial success, but it is a fantastic Arm Server platform that is supported upstream and runs perfect out-of-the-box. The SoftIron OverDrive 1000 comes in a small desktop style case, but the OverDrive 3000 series comes in a 1U chassis ready for rackmount installation. It has a BMC, 10gb ethernet, 14 SATA ports (!), and 2 PCIe slots. A standard UEFI boot process allows for easy installation of CentOS, RedHat, Debian, Ubuntu, SUSE, and any other Linux flavor that has an ARM64 build.  Though their Cortex A57 cores are getting a bit older now, they still make great build machines, especially when paired with fast SSD’s.

Next up is the Cavium ThunderX, and the newer ThunderX2. These chips are sold in servers from several vendors, which come in various shapes and sizes. Some of the examples we’ve found include the System76 Starling, the Avantek R-series in both 1U and 2U sizes, and the Gigabyte Arm offering that closely match Avantek’s specs. There are High Density designs, single processor and dual processor options, and 10 GBE as well as SFP options available.  ThunderX2’s have been more popular in HPC environments, but even a first-generation ThunderX is a great choice, and still a very powerful machine.  They can be purchased with up to 48-cores, or  in dual-processor configurations then containing up to 96 cores.

Another option is the Ampere eMag Arm Server from a company that formed a few  years ago, Ampere Computing.  They ship a turnkey Arm Server that is sold by Lenovo, the HR330A or the HR350A.  Their current-generation platform has 32 Arm cores running at 3.0ghz, 42 lanes of PCIe bandwidth, and 1 TB of memory capacity, and their next-generation product is said to have up to 80 Arm Neoverse N1 cores.  Current models are available for purchase from their website, or through Lenovo.

And of course, if buying physical servers and hosting them yourself, or placing them in a datacenter, is not feasible or cost effective in your situation, then our hosted Arm servers are a great option as well!  Our miniNodes Arm are certainly more modest in comparison to those mentioned above, but, they are a great way to get started with Arm development, testing existing code for compatibility, or lighter workloads that don’t require quite so much compute capability.

Be sure to check back often for all things Arm Server related!

Posted on

Arm Server Update, Summer 2019

It has been a while since our last Arm Server update, and as usual there has been a lot of changes, forward progress, and new developments in the ecosystem!

The enterprise Arm Server hardware is now mostly consolidated around the Cavium ThunderX2 and Ampere eMag products, available from Gigabyte, Avantek and Phoenics Electronics. Each can be purchased in 1U, 2U, and 4U configurations ready for the datacenter, and high performance developer workstations based on the same hardware are available, as well. Both of these solutions can be customized with additional RAM, storage, and networking, to best fit the intended workload.

Another option that exists, but is difficult to obtain in the United States, is the Huawei 1620, also known as the Kunpeng 920. These servers are also enterprise grade servers ready to be installed in a datacenter environment, typically in a 2U chassis with configurable memory and storage options. However, availability outside of Asia is limited, and new regulations may make importing them difficult.

While the Cavium, Ampere, and (potentially) Huawei servers are available as bare-metal options shipped directly to you for installation in your own datacenter, Amazon has also made significant progress over the past few months and is rapidly becoming the most popular Arm Server provider. They use their own Arm Server CPU called the Gravitron, that they use in their own proprietary AWS A1 ECS instances. This is quickly becoming the best way to deploy Arm Servers, as the entire system is in the Cloud and no hardware has to be purchased. They come in various sizes and price ranges, and experienced developers organizations who are familiar with the AWS system can simply pay by the hour for temporary workloads. For users who are less familiar with the ECS dashboard, less comfortable with the fluctuations in billing model, or prefer a fixed rate, we at miniNodes offer pre-configured Arm VPS servers in a range of sizes and prices, hosted atop the AWS platform.

Finally, the Edge of the network continues to be where a lot of innovation is occurring, and Arm Servers are a perfect fit for deplopyment as Edge Servers, due to their low power consumption, cost-effectiveness, and wide range of size and formats. The MacchiattoBin has been demonstrated running workloads in the base of windmills, the new SolidRun Clearfog ITX is promising to be a flexible solution, and the new Odroid N2 is an intersting device that has “enough” performance to satisfy a wide range of workloads that don’t need to always rely on the Cloud, and can instead deliver services and data to end-users (or other devices) faster by being located in closer proximity to where compute is needed.

As always, check back regularly for updates and Arm Server news, or follow us on Twitter where we share Arm related news on a daily basis!

Posted on

Report: Qualcomm Centriq TCO beats Intel

Tirias Research recently released a new Report detailing the Qualcomm Centriq Total Cost of Ownership versus an Intel Xeon x86 platform on a common workload, and the Qualcomm came out far ahead.  The full article is located here:  https://www.forbes.com/sites/tiriasresearch/2018/02/20/qdt-improved-server-tco/#3bbff2bc4675  The relevant piece is this:

Our TCO analysis demonstrated that using only one Qualcomm Centriq 2452 SoC per server chassis, a 12kW rack full of 36 46-core SoCs should show slightly better performance than a rack full of Intel Xeon Silver 4110 dual-socket server chassis, at only 51% of the power consumption. That’s similar performance with about half the power consumption.

 

Using two Qualcomm Centriq 2452 SoCs per server chassis in a 12kW rack should yield a little over double the performance of the dual-socket Intel Xeon Silver 4110 servers at 88% of the power consumption. A key factor is that only 35 of the Intel Xeon Silver 4110 systems can fit within the 12kW rack power budget. In this scenario, Qualcomm Centriq 2400 offers double the performance with less power consumption.

So, a single socket Centriq is essentially using half as much power for the exact same performance and workload, translating in to real savings.  And, there is room for performance improvement as well, by moving up to a dual socket design.  In that scenario, doubling the performance of the Xeon rack still results in a 12% power budget savings.  Double the performance and still drawing less power per rack, Qualcomm’s going to be challenging Intel’s dominance in the datacenter.

 

Posted on

ARM Server Update, Spring 2018

Continuing on with our quarterly updates to the ARM Server ecosystem, as usual there is quite a bit of news to report on!  Let’s dive right in to the analysis!

The Qualcomm Centriq continues to make headlines, with the first design win recently announced.  Hatch, a cloud gaming company, has chosen the Centriq 2400 to power it’s cloud gaming platform.  More information is available here:  https://www.forbes.com/sites/tiriasresearch/2018/02/20/hatch-qdt-cloud-gaming/

Qualcomm is also in the news for another reason as well.  Broadcom, another chip maker, has launched a hostile bid to takeover Qualcomm, although Qualcomm has thus far held off their unwanted pursuit, and is attempting to remain independent.  Consolidation in the chip maker space has been picking up in recent years, with the NXP purchase of Freescale, Intel buying Altera, Macom purchasing Applied Micro, and many more.

Which leads to the next news in the industry:  Macom had recently quietly sold off the Applied Micro assets to a secretively named buyer, known only as Project Denver Holdings.  However, they have now formed a new organization, called Ampere, who will continue on with the development and marketing of the X-Gene line of ARM Server processors.  More info on Ampere can be found here:  https://amperecomputing.com/

Finally, Linaro’s 96Boards team has brought to market a development workstation conforming to their Enterprise Edition standards.  The newly launched workstation features a 24-core Socionext Synquacer SoC, plus a hard drive, memory, and video card to round out the system.  It is currently listed for sale at $1,250, so it is not cheap, but it does fulfill a niched that has been underserved in the market.  It can be purchased here:  http://www.chip1stop.com/web/USA/en/search.do?dispPartIds=SOCI-0000001

 

Posted on

ARM Server Update, Fall 2016

Two major conferences devoted to the ARM ecosystem and technologies were recently held, ARMTechCon and Linaro Connect. Some new product announcements were revealed, and of course ARM Servers were front and center.

Linaro Connect featured the announcement and release of the new 96Boards IoT edition, a new smaller platform specifically designed for secure Internet of Things applications. There were also conference talks on the kernel, storage, Android, OCP, and more. But of course lots of attention was placed on the ARM Server updates, with the latest information on OpenStack, Xen, and processor technology announced. Linaro focuses on Linux on ARM of course, from both a hardware and software perspective.

ArmTechCon featured a more diverse set of topics, such as automotive, robotics, Internet of Things, and others. New application specific processors devoted to secure automotive and autonomous driving, network interconnects, and GPU’s were announced as well.

 

Posted on

ARM Server Linux Update, June 2016

As usual, a lot has changed in just a short time since our last update.  Here are some of the highlights from industry news.

First and foremost, the RaspberryPi 3 has continued to be the most popular ARM single board computer.  It now includes WiFi and Bluetooth, and the official Raspbian operating system has been upgraded to include support for the new features.  While it has a 64-bit processor, for the time being it still uses a 32-bit operating system.

Just a few days ago, we got some detail on the Cavium ThunderX2 processor that is forthcoming.  This is an enterprise-grade processor that will have 54 cores and support up to 100gb of ethernet bandwidth.  It will deliver 2x to 3x the performance of the current ThunderX processor, and should be able to compete head-to-head with Xeon’s in many workloads.

Finally, the Pine64 has been shipping in volume now, with most Kickstarter backers having received their boards.  The Pine64 is based on a 64-bit Allwinner A64 processor, which is not the most powerful around, but it sets a new low-price for 64-bit ARM hardware.  At just $15 for the entry level Pine64, the price of 64-bit ARM hardware has dropped from $3,000 to $15 in the course of about 1 year.  Talk about rapid innovation!

Posted on

Hosted Raspberry Pi 3 Servers Now Available!

miniNodes.com is proud to be the first cloud hosting provider to offer the new Raspberry Pi 3 as a hosted server.  The Raspberry Pi 3 combines a powerful new Broadcom quad-core 64-bit ARM processor, 1gb of RAM, and the reliable Raspbian Stretch linux operating system.  This makes the Raspberry Pi 3 a great platform for a small ARM server that offers plenty of compute capacity for basic services such as hosting a website or email, API hosting and development, lightweight development frameworks such as NodeJS application hosting, Internet of Things gateways and communication servers, IoT endpoints, Azure Edge container hosts, and more.  The Raspberry Pi 3 server is also a great way to experiment with ARM servers in the cloud, and ensure code compatibility with other more powerful ARM servers that are forthcoming.  Each hosted Raspberry Pi 3 server comes with SSH access and a dedicated IP address, making deployments to the server easy and familiar to developers.

Check them out here:  https://www.mininodes.com/product/raspberry-pi-3-server/

Posted on

How-to: Install Varnish and Nginx on Arm Servers

With the release of the Raspberry Pi 3 Model B, we read that Mythic Beasts was hosting a portion of the traffic going to the RaspberryPi.org site on an actual Pi 3! Pretty neat! We took a closer look and realized there wasn’t too much magic going on here, so we decided to write a quick how-to on setting up Varnish to cache content and speed up rendering and delivery of dynamic web pages. Since our little Arm servers are (obviously) not the most powerful platforms around, this can dramatically increase the performance and responsiveness of web servers running on Arm.

This guide will detail how to setup Varnish and Nginx on a Debian 8 “Jessie” Server running on Arm.

First and foremost, we need to perform the actual installation of the software, then we can configure it. We simply run:

sudo apt-get install nginx varnish.

Now, we need to make some changes to setup the environment properly.

The first file we need to take a look at is the Nginx configuration file which should be located at /etc/nginx/sites-available/default

Specifically, we need to move Nginx off of port 80 so that Varnish will be able to take over port 80 traffic. For this example, we will place Nginx on port 8080 instead, so, we update the file to reflect the following:

server
listen 8080; {
root /usr/share/nginx/html;
index index.html index.htm;

# Make site accessible from http://localhost/
server_name localhost.localdomain;

(We are not using a fully qualified hostname in this example, so later on we will go update our hosts file to reflect this.)

Next, we can configure Varnish to go look for Nginx on port 8080, by editing the Varnish config file located at /usr/local/etc/varnish/default.vcl and editing the ‘backend’ section. Here is a simple example: (Note – This one has some common WordPress login pages excluded from being cached)

vcl 4.0;
# Based on: https://github.com/mattiasgeniar/varnish-4.0-configuration-templates/blob/master/default.vcl

import std;
import directors;

backend default {
.host = “127.0.0.1”;
.port = “8080”;
}

sub vcl_recv {
# Do not cache following pages (edit as needed for your configuration).
if (!( req.url ~ “wp-(login|admin|comments-post)” )) {
return (pass);
}

# Drop the cookies for everything else
unset req.http.cookie;

# Try a cache-lookup
return (hash);
}

sub vcl_backend_response {
# Below will cache the page for one week.(1s = 1 sec, 1d = 1 day)
set beresp.ttl = 1w;
}

Once Varnish knows where to look for Nginx, we can now set Varnish to take over port 80 that we previously freed up. To do this, we need to modify the Varnish start file, which in this case is located at /etc/systemd/system/varnish.service because we are using Debian. If you are on another distro, this location may vary. We need to change the ‘-a’ flag from -a :6081 to -a :80, then save and close the file.

The last change we need to make is to our hosts file, to ensure that the local network is interpreted correctly. We need to edit the /etc/hosts file, and add a line that translates 127.0.0.1 to localhost and localhost.localdomain:

127.0.0.1 localhost localhost.localdomain

At this point, we could technically restart the Nginx and Varnish services to read in these changes, but a reboot is probably a good idea as well if you can afford it.

Once the server comes back up, you should now have Varnish running on port 80 caching content from your Nginx web server. One way to test this is to go to http://www.isvarnishworking.com, and put in your IP address (keep in mind, you need a public facing IP for this). If everything is working, you should see a message like this:

varnish-nginx-on-arm-server

Additionally, you can run ‘sudo varnishstat’ to see a detailed breakdown of how much caching is being done by Varnish.

So, thanks go to Raspberry Pi for inspiring us to attempt to duplicate their work. Hopefully this helps you install Varnish and Nginx on Arm, and speed up web page rendering on your Pi’s or other small Arm-based devices!