Amazon Web Services(AWS)
Cloud computing —
Cloud computing is an application-based software infrastructure that stores data on remote servers, which can be accessed through the internet. To understand how cloud computing works, it can be divided into front-end and backend.
The front end enables a user to access data stored in the cloud using an internet browser or a cloud computing software. However, the primary component of cloud computing — responsible for securely storing data and information — is the backend. It comprises servers, computers, databases, and central servers.
The central server facilitates operations by following a set of rules known as protocols. It uses a software, middleware, to ensure seamless connectivity between devices/computers linked via cloud computing. Cloud computing service providers usually maintain multiple copies of the data to mitigate instances of security threats, data loss, data breach, etc.
Types of Cloud Computing —
Public Cloud
Some public cloud examples include those offered by Amazon, Microsoft, or Google. These companies provide both services and infrastructure, which are shared by all customers. Public clouds typically have massive amounts of available space which translates into easy scalability. A public cloud is often recommended for software development and collaborative projects. Companies can design their applications to be portable, so that a project that’s tested in the public cloud can be moved to the private cloud for production. Most cloud providers package their computing resources as part of a service. Public cloud examples range from access to a completely virtualized infrastructure that provides little more than raw processing power and storage (Infrastructure as a Service, or IaaS) to specialized software programs that are easy to implement and use (Software as a Service, or SaaS).
The great advantage of a public cloud is its versatility and “pay as you go” structure that allows customers to provision more capacity on demand. On the downside, the essential infrastructure and operating system of the public cloud remain under full control of the cloud provider. Customers may continue to use the platform under the terms and conditions laid out by the provider, but they may have difficulty repatriating their assets if they want to change providers. Should the provider go out of business or make significant changes to the platform, customers could be forced to make significant infrastructure changes on short notice. There’s also the risk of an unpatched security vulnerability in the cloud architecture exposing customers to risk.
Private Cloud
Private clouds usually reside behind a firewall and are utilized by a single organization. A completely on-premises cloud may be the preferred solution for businesses with very tight regulatory requirements, though private clouds implemented through a colocation provider are gaining in popularity. Authorized users can access, utilize, and store data in the private cloud from anywhere, just like they could with a public cloud. The difference is that no one else can access or utilize those computing resources. Private cloud solutions offer both security and control, but these benefits come at a cost. The company that owns the cloud is responsible for both software and infrastructure, making this a less economical model than the public cloud.
The additional control offered by a private cloud makes it easier to restrict access to valuable assets and ensures that a company will be able to move its data and applications where it wants, whenever it wants. Furthermore, since the private cloud isn’t controlled by an outside vendor, there’s no risk of sudden changes disrupting the company’s entire infrastructure. A private cloud solution will also not be affected by a public cloud provider’s system downtime. But private clouds also lack the versatility of public clouds. They can only be expanded by adding more physical compute and storage capacity, making it difficult to scale operations quickly should the business need arise.
Hybrid Cloud
Hybrid clouds combine public clouds with private clouds. They are designed to allow the two platform to interact seamlessly, with data and applications moving smoothly from one to the other.
The primary advantage of a hybrid cloud model is its ability to provide the scalable computing power of a public cloud with the security and control of a private cloud. Data can be stored safely behind the firewalls and encryption protocols of the private cloud, then moved securely into a public cloud environment when needed. This is especially helpful in the age of big data analytics, when industries like healthcare must adhere to strict data privacy regulations while also using sophisticated algorithms powered by artificial intelligence (AI) to derive actionable insights from huge masses of unstructured data.
Advantages of Cloud Computing —
- Less Costs
The services are free from capital expenditure. There are no huge costs of hardware in cloud computing. You just have to pay as you operate it and enjoy the model based on your subscription plan. - 24 X 7 Availability
Most of the cloud providers are truly reliable in offering their services, with most of them maintaining an uptime of 99.9%. The workers can get onto the applications needed basically from anywhere. Some of the applications even function off-line. - Flexibility in Capacity
It offers flexible facility which could be turned off, up or down as per the circumstances of the user. For instance, a promotion of sales is very popular, capacity can be immediately and quickly added to it for the avoidance of losing sales and crashing servers. When those sales are done, the capacity can also be shrunk for the reduction of costs. - All over Functioning
Cloud computing offers yet another advantage of working from anywhere across the globe, as long as you have an internet connection. Even while using the critical cloud services that offer mobile apps, there is no limitation of the device used. - Automated Updates on Software
In cloud computing, the server suppliers regularly update your software including the updates on security, so that you do not need to agonize on wasting your crucial time on maintaining the system. You find extra time to focus on the important things like ‘How to grow your businesses. - Security
Cloud computing offers great security when any sensitive data has been lost. As the data is stored in the system, it can be easily accessed even if something happens to your computer. You can even remotely wipe out data from the lost machines for avoiding it getting in the wrong hands.
Case Study:
BMW Group with AWS —
The BMW Group is using AWS for its new connected-car application that collects sensor data from BMW 7 Series cars to give drivers dynamically updated map information. BMW Group is one of the leading manufacturers of premium cars and mobility services in the world, with brands such as Rolls Royce, BMW, and Mini. BMW built its new car-as-a-sensor (CARASSO) service in only six months leveraging Amazon Simple Storage Service (Amazon S3), Amazon Simple Queue Service (Amazon SQS), Amazon DynamoDB, Amazon Relational Database Service (Amazon RDS), and AWS Elastic Beanstalk. By running on AWS, CARASSO can adapt to rapidly changing load requirements that can scale up and down by two orders of magnitude within 24 hours. By 2018 CARASSO is expected to process data collected by a fleet of 100,000 vehicles traveling more than eight billion kilometers.
For predicting vehicle defects at BMW, a machine learning pipeline evaluating several thousand features was implemented. As important features can be useful for evaluating specific defects, a feature selection approach has been used. For further evaluating the importance of features, several feature selection techniques (filters and wrappers) have been implemented as ml PipelineStages for usage on dataframes for incorporation in a complete Spark ml Pipeline, including preprocessing and classification. The general steps for building custom Spark ml Estimators are presented. The API of the newly implemented feature selection techniques is demonstrated and results of a performance analysis are shown. Besides that, experiences gained and pitfalls that should be avoided are shared.
About BMW’s Unified Configurator Platform
BMW Group, the leading manufacturers of premium cars and provider of mobility services, used AWS to build its next-generation Unified Configurator Platform. At the AWS Summit Berlin 2018, BMW Group presented a deep dive into the journey from an on-premises, monolithic application to a microservices-based platform in the cloud.
When moving its Unified Configurator Platform into the cloud, BMW Group built a CI/CD pipeline based on Git, Jenkins and AWS CodePipeline, and used Amazon API Gateway, AWS Lambda and AWS Elastic Beanstalk to develop the platform’s microservices architecture. BMW Group furthermore leveraged AWS Professional Services and the Well-Architected Framework to ensure a fast and reliable project outcome, and worked with Game Day / Chaos Monkey methodology to train the operations team to monitor and maintain the infrastructure and applications.
Why AWS??
In the era of rapid technological change, the cloud is not only useful for backup and storage, it’s key for future-proofing businesses and maintaining a competitive edge. Businesses moving mass amounts of data to the cloud can take advantage of the latest technology and move away from aging on-premises infrastructure, putting themselves at the forefront of innovation.
BMW’s new connected-car application (CARASSO) is leveraging aws-s3, aws-sqs, aws-dynamodb, aws-rds, and aws-elastic beanstalk.CARASSO collects sensor data from BMW 7 Series cars to give drivers dynamically updated map information.
Disadvantages of AWS
Here, are drawbacks/cons of using AWS cloud:
- AWS deployment process is not easy and very lengthy which can take up to 15 to 20 minute for a simple website
- Unreliable deployment with no error description
- AWS is not an ideal option for start-ups that are not tech-savvy
- You need to deploy your application yourself using recipes, or manually.
- Launching multiple app instances is a very difficult process in AWS.
THANK YOU
References: