In this article, we will cover the following and more.<\/p>\n\n\n\n
- What is Cluster Computing on the Edge?<\/li>
- How does Clustering Really Work?<\/li>
- What are the Uses & Benefits of Clustering?<\/li>
- Hardware Recommendations for Cluster Computing on the Edge<\/li>
- Examples, Projects & Use Cases of Cluster Computing on the Edge<\/li><\/ul>\n\n\n\n<\/div>\n\n\n\n
![\"\"](\"https:\/\/blog.seeedstudio.com\/wp-content\/uploads\/2021\/04\/week8-1-1030x601.png\")
<\/figure><\/div>\n\n\n\n<\/div>\n\n\n\n
\n\n\n\n<\/div>\n\n\n\nWhat is Cluster Computing on the Edge?<\/strong><\/h2>\n\n\n\n
<\/div>\n\n\n\nWhen we mention Cluster Computing<\/em>, the first thing that might come to your mind is rows and rows of huge computers lined up on shelves, chugging along endlessly at intensive computing tasks. And you\u2019re right! By definition, computer clusters are groups of computers that work together<\/strong> so that they can be viewed as a single system.<\/p>\n\n\n\n
<\/div>\n\n\n\nSource: EasyTechJunkie<\/em><\/figcaption><\/figure><\/div>\n\n\n\n <\/div>\n\n\n\nHowever, today\u2019s article is about Cluster Computing<\/em> <\/strong>on the Edge<\/em><\/strong>. Now, what does \u201con the Edge\u201d mean?<\/p>\n\n\n\n
Edge Computing, or Computing on the Edge, refers to computational tasks or data processing being performed on devices which are physically located in the local space. This is in direct contrast to cloud computing, where these processes are handled remotely.<\/p>\n\n\n\n
Some examples of common edge devices are mobile phones, smart watches, and even automobile vehicles! Edge computing brings numerous benefits such as reduced latency, lower costs, increased data security and system reliability.<\/p>\n\n\n\n
You can also read more about edge computing in my previous article<\/a>!<\/p>\n\n\n\n
<\/div>\n\n\n\nSource: California Technical Academy<\/em><\/figcaption><\/figure><\/div>\n\n\n\n <\/div>\n\n\n\nTraditionally, cluster computing was unique to cloud computing, but as advances in Single Board Computers (SBCs) & network infrastructure take strides, this is no longer the case. For example, the Jetson Mate<\/a> Carrier Board shown below allows you to connect up to four Jetson Nano modules<\/a> to create your very own computer cluster in an extremely compact footprint of 11 x 12 cm!<\/p>\n\n\n\n
<\/div>\n\n\n\n<\/figure><\/div>\n\n\n\n
<\/div>\n\n\n\nIn simple terms, you can understand Cluster Computing on the Edge<\/em> as a new paradigm which aims to bring the benefits of cluster computing into edge computing<\/strong> to get the best of both worlds!<\/p>\n\n\n\n
<\/div>\n\n\n\n
\n\n\n\n<\/div>\n\n\n\nHow does Clustering Really Work?<\/strong><\/h2>\n\n\n\n
<\/div>\n\n\n\nWe know so far that a computer cluster consists of multiple machines, each of which are known as a node<\/strong>. In each cluster, there is usually a single head node<\/strong>, followed by multiple slave nodes<\/strong> (or worker nodes). All of them are connected to and able to communicate with each other, and typically run the same operating system.<\/p>\n\n\n\n
<\/figure><\/div>\n\n\n\n
When the cluster receives a job<\/strong>, such as a request to process and return some data, the head node is responsible for delegating the jobs to the slave nodes<\/strong>. The way that the computing workload is distributed is largely where clusters differ from each other. Some clusters, for instance, focus on delivering the highest performance possible, while others are designed to guard against failure.<\/p>\n\n\n\n
<\/div>\n\n\n\n
\n\n\n\n<\/div>\n\n\n\nWhat are the Uses & Benefits of Clustering at the Edge?<\/strong><\/h2>\n\n\n\n
<\/div>\n\n\n\nThere are three major types<\/strong> that you should know about. Each of them serve different purposes and contribute their own set of benefits to cluster computing!<\/p>\n\n\n\n
<\/div>\n\n\n\n- High Availability<\/strong> – Ensures that applications are always available by rerouting requests to another node in the event of a failure.<\/li>
- Load Balancing<\/strong> – Spreads computing workloads evenly across slave nodes to handle high job volumes.<\/li>
- High Performance<\/strong> – Multiple slave nodes are used in parallel to increase computing power for tasks with high computing requirements.<\/li><\/ul>\n\n\n\n
<\/div>\n\n\n\n1. Robust Edge Computing Systems<\/strong><\/h3>\n\n\n\n
<\/div>\n\n\n\nIn edge applications where low maintenance requirements are critical, clusters can bring greater system reliability. With a single computer, one failure would result in immediate system downtime. In a cluster, the multiple slave nodes build redundancies to guard against this, since other nodes can continue to function even if one fails. Repairing hardware faults in clusters is also easier, since you only have to reconfigure a single node instead of the entire infrastructure.<\/p>\n\n\n\n
<\/div>\n\n\n\n2. Flexible & Suitable for a Wide Range of Edge Applications<\/strong><\/h3>\n\n\n\n
<\/div>\n\n\n\nFrom the various types of clusters, it\u2019s easy to see that cluster computing has a place in many different types of applications. For example, a high performance cluster can be used to run machine learning on the edge, whereas a high availability cluster will be suitable for hosting servers to process and report data from IoT devices. Furthermore, clustering paradigms are software based – meaning that the same hardware can be easily reconfigured to meet the needs of another application!<\/p>\n\n\n\n
<\/div>\n\n\n\n3. Greater Computing Power on the Edge<\/strong><\/h3>\n\n\n\n
<\/div>\n\n\n\nWith a cluster, you can easily add more modules according to your computing requirements. For example, a cluster can have up to dozens of multiprocessor machines, delivering far greater power than a single computer! This drastically increases the possibilities of edge computing, since edge computing devices have traditionally offered less computational capacity.<\/p>\n\n\n\n
<\/div>\n\n\n\n
\n\n\n\n<\/div>\n\n\n\nHardware Recommendations for Cluster Computing on the Edge<\/strong><\/h2>\n\n\n\n
<\/div>\n\n\n\nIf you\u2019re keen to get started with cluster computing in edge applications, it\u2019s a good idea to get a feel for the hardware available. In this section, I\u2019ll go through some of my top recommendations for your consideration!<\/p>\n\n\n\n
<\/div>\n\n\n\nRaspberry Pi 4<\/strong><\/a> \/ <\/strong>Compute Module 4<\/strong><\/a><\/h3>\n\n\n\n
<\/div>\n\n\n\nThe Raspberry Pi 4 is arguably the most popular Single Board Computer, offering decent computing power and rich I\/O capabilities in a credit card sized form factor. Running a Linux operating system and with extensive documentation and community support, the Raspberry Pi 4 (and compute module) is a great option for building your very first computer cluster! In this official tutorial<\/a>, a Raspberry Pi Cluster is used to speed up calculations by a factor of nearly five!<\/p>\n\n\n\n
<\/div>\n\n\n\n<\/figure><\/div>\n\n\n\n
<\/div>\n\n\n\nProduct Features:<\/strong><\/p>\n\n\n\n
- Broadcom BCM2711, quad-core Cortex-A72 (ARM v8) 64-bit SoC @ 1.5GHz<\/li>
- 1GB, 2GB or 4GB LPDDR4 RAM<\/li>
- 2.4 GHz and 5.0 GHz IEEE 802.11b\/g\/n\/ac wireless LAN, Bluetooth 5.0, BLE, Gigabit Ethernet<\/li>
- 2 \u00d7 USB 3.0 ports \/ 2 \u00d7 USB 2.0 ports<\/li>
- Standard 40-pin GPIO Header<\/li>
- 2 \u00d7 micro HDMI ports (up to 4Kp60 supported)<\/li>
- 2-lane MIPI DSI display port \/ 2-lane MIPI CSI camera port<\/li><\/ul>\n\n\n\n<\/div>\n\n\n\n
Interested to learn more? Learn more about the Raspberry Pi 4 on the Seeed Online Store<\/a> now!<\/p>\n\n\n\n
<\/div>\n\n\n\nODYSSEY X86J4105800<\/strong><\/a><\/h3>\n\n\n\n
<\/div>\n\n\n\nRunning on the x86 CPU architecture, the ODYSSEY x86 is a powerful single board computer in its own right. With a cluster, you can extend these possibilities even further with much greater computing capabilities! With a wealth of I\/O interfaces, the ODYSSEY x86 will definitely serve you well for any kind of edge application.<\/p>\n\n\n\n
<\/div>\n\n\n\n<\/figure><\/div>\n\n\n\n
<\/div>\n\n\n\nProduct Features:<\/strong><\/p>\n\n\n\n
- Intel\u00ae Celeron\u00ae J4105, Quad-Core 1.5-2.5GHZ<\/li>
- Dual-Band Frequency 2.4GHz\/5GHz WiFi<\/li>
- Intel\u00ae UHD Graphics 600<\/li>
- Dual Gigabit Ethernet<\/li>
- Integrated Arduino Coprocessor ATSAMD21 ARM\u00ae Cortex\u00ae-M0+<\/li>
- Raspberry Pi 40-Pin Compatible<\/li>
- 2 x M.2 PCIe (B Key and M Key)<\/li>
- Support Windows 10 & Linux OS<\/li>
- Compatible with Grove Ecosystem<\/li><\/ul>\n\n\n\n<\/div>\n\n\n\n
Interested to learn more? Learn more about the ODYSSEY x86J4105800 on the Seeed Online Store<\/a> now!<\/p>\n\n\n\n
<\/div>\n\n\n\nNVIDIA Jetson Nano Module<\/strong><\/a><\/h3>\n\n\n\n
<\/div>\n\n\n\nIf you\u2019re intending to run machine learning workloads on the edge, you should definitely consider the NVIDIA Jetson Nano Module. Designed specially for AI applications with NVIDIA\u2019s JetPack SDK<\/a>, you can easily build, deploy and manage powerful machine learning at the edge with low power consumption. The Jetson Nano can be paired with the Jetson Mate<\/a> to easily build powerful clusters!<\/p>\n\n\n\n
<\/div>\n\n\n\n<\/figure><\/div>\n\n\n\n
<\/div>\n\n\n\nProduct Features<\/strong>:<\/p>\n\n\n\n
- Quad-Core ARM Cortex-A57 MPCore Processor<\/li>
- NVIDIA Maxwell GPU with 128 NVIDIA CUDA Cores<\/li>
- 4GB 64-Bit LPDDR4 Memory at 1600MHz 25.6GBps<\/li>
- 16GB eMMC Storage<\/li>
- NVIDIA JetPack SDK for AI Development<\/li><\/ul>\n\n\n\n<\/div>\n\n\n\n
Pick up your very own NVIDIA Jetson Nano Developer Kit on the Seeed Online Store<\/a>!<\/p>\n\n\n\n
<\/div>\n\n\n\n
\n\n\n\n<\/div>\n\n\n\nCluster Computing Examples: Projects & Use Cases<\/strong><\/h2>\n\n\n\n
<\/div>\n\n\n\nComputer clusters can be used just about in any situation that you can imagine, providing improved performance and reliability in comparison to traditional single device computing. Here are a few examples where cluster computing on the edge can be especially useful!<\/p>\n\n\n\n
<\/div>\n\n\n\nMachine Learning Inference Server<\/strong><\/h3>\n\n\n\n
Run a GPU cluster to run rapid calculations required for machine learning! This is similar to cloud computing, except we bring computing capabilities much closer to where they are needed on the edge. For example, you can use a cluster of Jetson Nano modules and use multiple GPUs simultaneously to run low latency, real time<\/strong> person detection on multiple video feeds in security systems!<\/p>\n\n\n\n
<\/div>\n\n\n\nSource: TowardsDataScience<\/em><\/figcaption><\/figure><\/div>\n\n\n\n <\/div>\n\n\n\nSelf-Hosted Private Cloud Server<\/strong><\/h3>\n\n\n\n
<\/div>\n\n\n\nYou can already do this with a single SBC setup, but deploying your own web server on a cluster allows you to scale this project up almost indefinitely<\/strong> – even to an enterprise level! This means you\u2019ll be able to host more data and applications to meet any kind of needs that you or your organisation might have.<\/p>\n\n\n\n
This is also a far more cost effective method compared to the traditional full server rack, since the hardware costs are greatly mitigated by affordable edge computing devices! In addition, enjoy the added security of your very own internal network!<\/p>\n\n\n\n
<\/div>\n\n\n\nTest & Deploy Applications with Kubernetes<\/strong><\/h3>\n\n\n\n
<\/div>\n\n\n\nIt\u2019s definitely worth mentioning that you can use Kubernetes to efficiently test and deploy any kind of application on your edge clusters. Kubernetes<\/a> is an open-source platform for managing containerised workloads and services, and is one of the most popular ways to deploy computer clusters in 2021. You can think of it as a management interface that helps you manage your clusters, scaling resources up or down as required to make the most efficient use of your hardware and more!<\/p>\n\n\n\n
<\/div>\n\n\n\nSource: Kubernetes<\/em><\/figcaption><\/figure><\/div>\n\n\n\n <\/div>\n\n\n\n
\n\n\n\n<\/div>\n\n\n\nTutorial: Build a Kubernetes Cluster with Jetson Mate<\/strong><\/h2>\n\n\n\n
<\/div>\n\n\n\nIn this section, I\u2019m going to show you just how easy it is to set up your very own cluster with the Jetson Mate and the powerful Jetson Nano modules. You can also read the complete tutorial on our Seeed Wiki page<\/a>.<\/p>\n\n\n\n
<\/div>\n\n\n\nRequired Materials<\/strong><\/h3>\n\n\n\n
<\/div>\n\n\n\nTo follow along with this tutorial, the following items are recommended. Take note that you will need to have at least two Jetson Nano modules, since we require a minimum of one master \/ head node and one worker \/ slave node.<\/p>\n\n\n\n
- Jetson Mate Carrier Board<\/a><\/li>
- Jetson Nano Module<\/a> (at least 2)<\/li>
- Qualified Type-C Power Adapter (65W minimum) with PD Protocol<\/li><\/ul>\n\n\n\n
<\/div>\n\n\n\nInstall & Configure Jetson OS<\/strong><\/h3>\n\n\n\n
<\/div>\n\n\n\nWe will have to configure the operating system for each of the modules using NVIDIA\u2019s official SDK manager. First, choose the target hardware as shown below.<\/p>\n\n\n\n
<\/div>\n\n\n\n<\/figure><\/div>\n\n\n\n
<\/div>\n\n\n\nThen, choose the OS and Libraries you want to install:<\/p>\n\n\n\n
<\/div>\n\n\n\n<\/figure><\/div>\n\n\n\n
<\/div>\n\n\n\nDownload and install the files. While downloading, insert the Jetson Nano compute module into the main node of the Jetson Mate.<\/p>\n\n\n\n
<\/div>\n\n\n\n<\/figure><\/div>\n\n\n\n
<\/div>\n\n\n\nFollowing this, short the 2 GND pins <\/strong>according to the picture shown.<\/p>\n\n\n\n
<\/div>\n\n\n\n<\/figure><\/div>\n\n\n\n
<\/div>\n\n\n\nConnect the Jetson Mate to your computer via the micro USB port and power on the machine by pressing the wake up <\/strong>button.<\/p>\n\n\n\n
The final step is to flash the operating system onto the compute module. When the installation of the OS and software library is completed, you will see a window pop up. Select Manual Setup option, then click flash and wait until completion. That\u2019s it!<\/p>\n\n\n\n
<\/div>\n\n\n\n<\/figure><\/div>\n\n\n\n
<\/div>\n\n\n\nTake note that all the modules can only be flashed when installed on the main node<\/strong>. You are required to flash and configure all your modules one by one on the main node.<\/p>\n\n\n\n
<\/div>\n\n\n\nRunning Kubernetes on our Cluster<\/strong><\/h3>\n\n\n\n
<\/div>\n\n\n\nIn the following steps, we will install and configure Kubernetes to run on our cluster of NVIDIA Jetson Nano modules!<\/p>\n\n\n\n
<\/div>\n\n\n\nConfiguring Docker<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\nFor both Worker & Master modules, <\/em><\/strong>we need to configure the docker runtime to use “nvidia” as default. <\/p>\n\n\n\n
Modify the file located at \/etc\/docker\/daemon.json as follows.<\/p>\n\n\n\n
<\/div>\n\n\n\n{\n \"default-runtime\" : \"nvidia\",\n \"runtimes\": {\n \"nvidia\": {\n \"path\": \"nvidia-container-runtime\",\n \"runtimeArgs\": []\n }\n }\n}<\/code><\/pre>\n\n\n\n<\/div>\n\n\n\nRestart the Docker daemon with the following command,<\/p>\n\n\n\n
<\/div>\n\n\n\nsudo<\/strong> systemctl daemon-reload && sudo systemctl restart docker<\/code><\/pre>\n\n\n\n<\/div>\n\n\n\nthen validate the Docker default runtime as NVIDIA.<\/p>\n\n\n\n
<\/div>\n\n\n\nsudo<\/strong> docker info | grep -i runtime<\/code><\/pre>\n\n\n\n<\/div>\n\n\n\nHere\u2019s a sample output:<\/p>\n\n\n\n
<\/div>\n\n\n\nRuntimes: nvidia runc\nDefault Runtime: nvidia<\/code><\/pre>\n\n\n\n<\/div>\n\n\n\nInstalling Kubernetes<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n - Jetson Nano Module<\/a> (at least 2)<\/li>
- Jetson Mate Carrier Board<\/a><\/li>
- Load Balancing<\/strong> – Spreads computing workloads evenly across slave nodes to handle high job volumes.<\/li>
- High Availability<\/strong> – Ensures that applications are always available by rerouting requests to another node in the event of a failure.<\/li>
![\"\"](\"https:\/\/lh3.googleusercontent.com\/l6NscuvIPOB4IcOx9UWDGpTzziEwc6oYv99inXle7Ga-MQ_kTlYsm1k0jlMElo6_Rq3pWgMnvrDa2PO9nGnbR029gYF9DMooJ2XY8l2SeGtiLmEgANgdNMfcvZEAr9B4RJ5jUHyG\")
![\"\"](\"https:\/\/lh3.googleusercontent.com\/VWl7dM1uWnfqyKOJUbWMr0mfNcB0ZOg86eRBFbsCWIlC6P8Dbd78b4O1f1Rlhod2AJxe9cGPlhR_ZTXusDtPWXJWM9TGmzj3qMqBAuIFEQuWaguIY3_7NRf4l1AzoQykmhDr8dEU\")
![\"\"](\"https:\/\/lh4.googleusercontent.com\/L9McFYlp5ew2OqeQ8K299gWBsd1vKh2hLpcZT2vxcl1gf3qs2GwnxS4GpyUD4VCKenNfe0T7d9nv-4EAWbnzzZn2X1guNutPxYLfHoHuW0bSxjGZpbKDeGjTA65ROV0bAU-8UIza\")
![\"\"](\"https:\/\/lh3.googleusercontent.com\/UdOA4NU4H-ncT68MS0HkLJyJ7F8ptRw36RXxNUd6fugpYKaILZIZ6LyUuJJW192mp_UWXoc87-q47vYEP53KNW7HuHUVJ2TEX-KWOfTlc7KrjJj1crMvbN6tqcQ6kYSvNDirMIqi\")