Bright Cluster Manager lets you administer clusters as a single entity, provisioning the hardware, operating system, and workload manager from a unified interface. This makes it easier to build a reliable cluster. Once your cluster is up and running, the Bright cluster management daemon keeps an eye on virtually every aspect of every node, and reports any problems it detects in the software or the hardware, so that you can take action and keep your cluster healthy. The intuitive Bright management console lets you see and respond to what's going on in your cluster anywhere, anytime.
Introduction
ASA Computers is now a partner of Bright Computing to offer a Top Linux cluster automation and management software, for HPC and machine learning to the market.Bright Computing is a global leader in Linux cluster automation and management software, for HPC and machine learning. Bright Cluster Manager™ automates the process of installing, provisioning, configuring, managing, and monitoring clusters for HPC, data analytics and machine learning environments. Bright's products are deployed in data centers around the world for organizations in healthcare, manufacturing, oil and gas, energy, pharmaceutical, financial, academic, and government. Bright has reseller agreements and technology partnerships with leading enterprise IT providers, including Dell/EMC, Hewlett Packard Enterprise, Cray, IBM, Intel, Huawei, Fujitsu, Nvidia, Microsoft, and Amazon. www.brightcomputing.com.
WHY BRIGHT COMPUTING
Built-in expertise
Task automation based on more than a decade of real-world use
Integrated Build, Management & Monitoring
Everything in one package with one interface
Platform Independent
Mix and match virtually any vendor hardware. Host HPC, machine learning, and analytics apps on the same cluster
Proven
Thousands of successful, growing installations
Future-proof
A continuous stream of new functionality ensures that our software is always one step ahead of where you're going
Commercial Support
We've earned a reputation for providing world class support, just ask our clients
Bright Computing Solutions
CLUSTERS FOR EDGE COMPUTING
Edge computing allows a business to place infrastructure in remote locations that lack IT staff to deliver new applications that take advantage of locally generated data to improve operations, services, experiences and products.
Bright Edge is a feature of Bright Cluster Manager that allows organizations to deploy and centrally manage computing resources in distributed locations as a single clustered infrastructure, from a single interface. The distributed computing nodes deployed and managed by Bright Cluster Manager can be imaged to support any workload, re-imaged on the fly to support different workloads when desired, and are monitored to ensure that you always know precisely what's going on. And when you need to add more computing capacity at any location, bring additional nodes online is quick and easy.
CLUSTERS FOR HYBRID CLOUD
Linux clusters deployed on-premise and Linux clusters in the cloud each offer advantages and disadvantages leading many organizations to pursue a hybrid approach that includes both. Building an effective hybrid cluster involves the daunting task of building the on-premise cluster and then enabling that cluster to seamlessly extend to the public cloud for additional resources when needed. Given the latency between on-premise and cloud, the goal should be to have a cluster with resources in your datacenter and in the cloud — each servicing local workloads – yet centrally managed as a single logical entity. Administratively combining these different and disparate environments is incredibly complex and can lead to costly consequences if cloud resources are not provisioned and terminated in a timely manner. Bright software automates the process of building and managing your Linux cluster, and also automates the process of extending your on-premise cluster to the public cloud. After building and deploying your on-premise cluster, a wizard guides you through the process of enabling your cluster to extend to either AWS or Azure for additional resources.
CLUSTERS FOR MACHINE LEARNING
The promise of machine learning is pulling every organization in every industry into the realm of high-performance computing. Many organizations have little or no experience with these systems, and even the most experienced high-performance computing practitioners will tell you that building and managing high-performance Linux clusters is no easy task. With hundreds or thousands of hardware and software elements that must work in unison, spanning compute, networking and storage, the skills, knowledge and experience required to do this is often more than an organization can cope with. Bright Cluster Manager software offers an integrated solution for building and managing machine learning clusters that reduces complexity, accelerates time to value and provides enormous flexibility. Bright Cluster Manager also provides a pre-tested catalog of popular machine learning frameworks and libraries, as well as integration with Jupyter Notebook, to ensure that data scientists can be as productive as possible and not waste time managing their work environment.
Bright Computing Products
BRIGHT CLUSTER MANAGER
Advanced Cluster Management Made EasyFeatures & Benefits
- Installation Wizard - Install everything you need, including your chosen distribution of Linux, workload manager, libraries, and more.
- Node Provisioning - Handle the provisioning of software images to nodes with ease, for clusters of virtually any size and complexity.
- Graphical User Interface - A single, intuitive interface for all cluster management functionality.
- Comprehensive Monitoring - Monitor, visualize, and analyze a comprehensive set of hardware and software metrics with ease.
- GPU Management - Leverage powerful GPU management and monitoring capabilities to gain maximum control.
- Cloudbursting - Easily create new clusters in the cloud, or add cloud-based resources to your existing clusters on-the-fly.
Deploys Easily
Builds your cluster from bare metal servers or virtual machines (VMs) with nothing to pre-install. Installs everything you need, including HPC workload managers, Kubernetes, Spark and Deep Learning libraries on your chosen distribution of Linux—Red Hat Enterprise Linux, CentOS, SUSE Linux Enterprise Server, or Ubuntu
Automates and Pre-Checks Everything
Automatically sets up networking, user directories, DNS and firewalls, then runs automated tests to ensure that everything works properly
Supports Mixed Environments
Simplifies management of multiple-architecture and multipleOS clusters by combining different types of servers, architectures, and operating systems in the same cluster
Provides Comprehensive Monitoring
Monitors, visualizes, and analyzes a comprehensive set of hardware, software, and job-based metrics using workload accounting and reporting features
Integrates with the Cloud
Automatically provisions additional resources from AWS, Azure and Open Stack based clouds when needed
Manages Accelerators (GPUs, FPGAs, IPUs)
Tracks and reports status and utilization of accelerators on a per-job basis to ensure efficient utilization by end users
Optimizes Use of Cluster Resources
Ensures resources are allocated to applications according to demand and based on policies you define.
Supports Containers
Provides the prerequisite underlying cluster for running Kubernetes, Docker and Singularity, and manages these packages through the Bright management interface
Includes Two Powerful User Interfaces
Provides two user interfaces so you can choose to provision, monitor, and manage your clusters with a traditional command line interface or the Bright View web-based graphical user interface
Easily Build and Manage Clusters for HPC, Deep Learning, Edge Computing, and More
Bright Cluster Manager lets you quickly build complete clusters from bare metal in minutes and manage them effectively. Bright combines provisioning, monitoring and change management facilities on a single tool that spans the entire lifecycle of your cluster, allowing administrators to provide better support to end users and your business
Quickly Deploy a Cluster for HPC, Deep Learning, or Edge Computing
When you need to get a cluster up and running, don't waste time cobbling together a solution from disparate open source tools. Bright software comes with everything you need to set up a complete cluster from bare metal and manage it with a powerful user interface. Whether your cluster is on-premises or in the cloud, Bright is the right choice for all your clustered infrastructure projects.
Dynamic Clusters with In-depth Accounting and Reporting
Bright's dynamic cloud provisioning capability enables you to build an entire cluster in the cloud or expand your physical cluster into the cloud for extra capacity. Bright can provision compute resources in Amazon Web Services or Microsoft Azure automatically and on-demand. With Workload Accounting and Reporting, you can isolate resource utilization by project and limit the visibility of resource usage to the owner of each project
Manage Edge Computing Infrastructure
The Bright Edge feature in Bright Cluster Manager allows you to deploy, manage and monitor geographically distributed cluster resources, including edge compute servers located in remote areas that lack skilled IT resources.
Proven Technology
Bright Computing has been building enterprisegrade cluster management software for more than a decade. Bright Cluster Manager has been successfully deployed in thousands of locations around the globe, providing an unparalleled level of maturity know-how and battle-tested software that you can count on.
Clusters in the Cloud
Bright Cluster Manager can provision and manage clusters that run either on-premises or in a public cloud. You can build an entire cluster in Amazon Web Services (AWS) or Azure from scratch, or extend a physical cluster into the cloud when you need extra capacity. Use Bright's new Auto-Scaling feature to automatically scale HPC queues and Kubernetes applications from pools of both local nodes and cloud nodes to create a highly effective hybrid cloud environment.
Bright Cluster Manager for Data Science
Deep Learning Infrastructure Management Made EasyBright Cluster Manager for Data Science lets you administer data science clusters as a single entity, provisioning the hardware, operating system, big data and deep learning software from a single interface. This makes it easier to build a reliable, clustered data science infrastructure. Once your system is up and running, the Bright cluster management daemon keeps an eye on virtually every aspect of every node, and reports any problems it detects in the software or the hardware so that you can take action and keep your infrastructure healthy. The intuitive Bright management interface makes it easy to see and take action on any issue impacting data science cluster performance.
Features & Benefits
- Installation Wizard - Install everything you need, including your chosen distribution of Linux, workload manager, libraries, and more.
- Flexible Configuration - Bright's configuration groups make Spark configuration more flexible than ever
- Graphical User Interface - A single, intuitive interface for all cluster management functionality.
- Choice of Machine Learning Libraries and Frameworks - Includes popular machine learning libraries and frameworks such as NVIDIA CUDA Deep Neural Network library (cuDNN), Chainer, MXNet, PYTorch, TensorFlow, Theano, OpenCV, XGBoost, and more.
- Supports Alternate File Systems - Take advantage of Lustre as an alternative to HDFS.
- Cloud - Easily create new clusters in the cloud, or add cloud-based resources to your existing clusters on-the-fly.
Modern Deep Learning Environment
Provides everything you need to quickly spin up a complete machine learning environment for data scientists that is cluster-ready and easy to manage
Choice of Machine Learning Libraries and Frameworks
Includes the most popular machine learning libraries and frameworks such as NVIDIA cuDNN, CUB, CUDA, TensorRT, Dynet, Fastai, JupyterHub, CCL2, MXNet, pyTorch, Chainer, CNTK, Horovod, Keras, Opencv3, Protobuf3, Torch, Tensorflow, Theano, and XGBoost (see: https://support.brightcomputing.com/ packages-dashboard/).
Scale-out JupyterHub
JupyterHub makes data science easy to use. Bright provides custom JupyterHub spawners that allow notebooks to be scheduled through the HPC scheduler or Kubernetes
Support for NVIDIA GPU Cloud (NGC)
Makes it easy to use NVIDIA NGC deep learning containers in so many ways—run them in containers, through a batch scheduler, on physical nodes, or in a cloud Enhances job metrics, accounting and reporting functionality making it possible to determine GPU usage on a per-job basis over time
Simple Deployment Process
Installs all the software and creates the necessary configuration files by simply answering a few questions about your cluster and Bright takes care of the rest Builds your new cluster right the first time and every time—on bare metal or in the cloud.
Bright's user interface makes it easy to monitor and manage your data science clusters.
Customizable dashboards help administrators ensure all nodes in a cluster are optimized for the workloads they run.
Simple deployment of deep learning infrastructure
With Bright, you can deploy deep learning environments in minutes. Bright simplifies management of multiple-architecture and multipleOS clusters by combining different types of servers, architectures, and OSes in the same cluster.
Developer-friendly environment
Bright for Data Science provides and installs environment modules that make it easy to dynamically modify the user environment. The software comes with a complete set of tools and libraries for HPC, Deep Learning, and Apache Spark so you will be ready to develop, debug, and deploy your code immediately.
Scalability
Bright makes HPC technology available to machine learning frameworks, allowing deep learning workloads to be spread across multiple machines. Bright's new Auto-Scale feature automatically scales HPC queues and Kubernetes applications from pools of both local nodes and cloud nodes to create a highly effective hybrid cloud environment.
Extended capabilities
Need more capacity? Extend your GPU-enabled instances into the cloud using Bright's cloud bursting capability. Bright also makes it easy to use NVIDIA machine learning containers (NGC) in your own private OpenStack cloud. In addition, Bright OpenStack includes the Stein release of OpenStack and the Nautilus release of Ceph to provide multi-attach for RDB volumes, optional deferred asynchronous volume deletion, bandwidthaware VM scheduling, and more
Bright OpenStack 9.0
Deep Learning Infrastructure Management Made EasyBright Computing has been providing management solutions for clustered systems for many years, and we've brought that experience to OpenStack management. Bright lets you administer your private cloud infrastructure as a single entity, provisioning the hardware, operating system, and cloud framework from a single interface. This makes it easier to build a robust private cloud, reliably, with the resources you already have. Once your cloud is up and running, our daemon keeps an eye on virtually every aspect of every node, and reports any problems it detects in the software or the hardware so that you can take action and keep your infrastructure healthy. The Bright management interface lets you see what's going on from top to bottom.
Features & Benefits
- Simple Deployment - The Bright installation wizard makes it easy. Just answer a few questions and Bright takes care of the rest.
- Intuitive Management GUI - Cluster Management GUI (CMGUI) is optimized for performing daily OpenStack management operations
- Configure using Role Assignments - Bright creates configuration files automatically using roles, so you won't have to manage configuration files manually.
- Integrated Monitoring and Health Checking - Bright monitors the OpenStack software, the servers, the network, and the operating system so you can rest assured your cloud is running as it should.
- Headache-free Networking Setup - Bright cuts through the complexity of OpenStack networking setup so you can have a working configuration quickly.
- Manage Storage Using Ceph - Ceph allows you to pool all of your server storage into a reliable redundant storage system. This enables you to scale with ease, and replace hardware without disruption.
Automate Deployment
Simply answer a few questions about the cloud you want to build, and Bright OpenStack takes over from there. In addition to installing all the software you need and creating the necessary configuration files, Bright OpenStack simplifies management of multiple-architecture and multiple-OS clusters by combining different types of servers, architectures, and OSes in the same cluster.
Install on Bare Metal
With Bright, there is nothing to pre-install. You can start with bare metal servers or virtual machines (VMs), and Bright OpenStack will install and configure everything you need.
Configure OpenStack Using Role-based Assignments
Bright OpenStack creates and automatically updates configuration files using Bright roles, so you don't have to manage configuration files manually.
Integrate Monitoring and Health Checking
Bright OpenStack keeps an eye on the OpenStack software, servers, network, and OS, so you know your cloud is running efficiently. In addition, Workload Accounting and Reporting makes it possible to isolate resource utilization by project and limit visibility of resource usage to the owner of the project.
Simplify Networking Setup
Bright cuts through the complexity of OpenStack networking setup so you can have a working configuration on the first try.
Manage Storage Using Ceph
In addition to the Stein release of OpenStack, Bright OpenStack also includes the Nautilus release of Ceph to provide important features such as multi-attach for RDB volumes, optional deferred asynchronous volume deletion, bandwidth-aware VM scheduling, and a new OpenStack placement service.
Bright View is a state-of-the-art user interface that simplifies monitoring and managing your OpenStack cloud.
Customizable dashboards help administrators ensure all nodes in a cluster are optimized for the workloads they run.
Quick to deploy. Easy to manage.
Bright takes care of all the tricky details when installing and configuring your cloud's operating system, networking, and OpenStack software. Once your cloud is up and running, Bright OpenStack's elegant and easy-to-use interface provides a powerful dashboard to monitor and manage the cloud throughout its lifecycle.
Proven Technology
Bright has been building management software for more than a decade. Our solutions are deployed in thousands of locations around the globe. Why struggle to build out your cloud infrastructure when you can rely on the experience built into Bright software? .
Easy Payment Options
Amazon AWS and Microsoft Azure are now integrated with Bright, so usage of Bright in the cloud is billed on an hourly basis and included on the monthly statement from your cloud provider.
Quickly Deploy a Private Cloud
Switching your data center from discrete services to a cloud can help you get the most from your equipment. Bright's complete solution makes it easy to build a orking cloud, optimizing the utilization of the servers in your data center, while also maintaining the service levels your users demand.
Speed Up DevOps
Bright OpenStack gives you the ability to stand up and tear down a cluster in whatever configuration you need, whenever you need it. This advanced yet easy-touse solution puts considerable computing power in the hands of your DevOps team, while eliminating the time-consuming and frustrating task of building clusters the old-fashioned way. Your team can simply decide how many nodes they need, choose the type the VMs to use, select a Linux distribution, and press ENTER. That's all there is to it. Each cluster created using “Cluster on Demand” has its own self-contained instance of Bright Cluster Manager to ensure easy management and monitoring.
Request Information
To speak with one of our Bright Computing Software expert please call us at +1 650-230-8000.You can also email us at 'sales@asacomputers.com' or simply click on the 'Request Info' button. Request Info