GCP Products Introduction
GCP vs On-premise IT Infrastructure: What Are the Differences?
Traditional On-premise IT Infrastructure
Costly and time-consuming for server room construction
Unable to expand the resources on-demand
Higher upfront capital investment
Higher cost for manual operating system & software installation
Manual IT infrastructure setup
Additional cost for hardware maintenance
Slow crisis response at data center and high risk of data loss
Redesign of API is needed for new projects
Google Cloud Platform
Virtual servers that provision in seconds
Auto scaling according to your concurrent loading
Only pay for the amount that you have used
Auto provision operating systems & software in a short period of time at lower cost
Quick deployment of complete IT Infrastructure
No hardware maintenance is required
Quick crisis response and complete data backup
Get started with new projects in a short period of time using the Google APIs
Main services of Google Cloud Platform (GCP)
Google Cloud is providing different functions based on three types of services – IaaS*, PaaS* and SaaS*. The followings are the main services provided by GCP.
The Compute Engine provided by Google provides high-performance and scalable Virtual Machines (VMs), which is one of the IaaS (Infrastructure as a Service) that able to avoid additional operation, staffing and expansion cost compared to hosting servers at local server rooms or data centers.
The VMs boot quickly and provide stable & high performance.
Virtual Machines are available in many instance types, or you can customize by your specific needs. GCP’s per-second billing and Committed Use Discounts helps you save cost.
App Engine is Google’s fully managed serverless application platform classified as PaaS.
App Engine enables developers and engineers to build and deploy applications by using many of the popular programming languages like Java, PHP, Node.js, Python, C#, .Net, Ruby, etc.
The advantage is that developers do not have to build their own environment or manage the server. In addition, it is common that the existing hardware machine is not able to handle the software expansion during the development stage, but the App Engine will automatically up and down scale so that developers are free from this problem.
Google Kubernetes Engine (GKE) is a managed, production-ready environment for deploying containerized applications. Container is a lightweight virtual environment that differ from VM, forming the concept of ‘Microservices’. Operating System installation is needed on VM for the running of application. Containers technology allows the virtual environments to share a common operating system. Multiple application environments can be deployed rapidly without installation operating systems individually for every single application.
How To Manage a Huge Container Cluster?
Google launched GKE in 2015 to solve the problem. Users are able to enjoy managed service provided by Google and deploy applications within the container for both stateless and stateful services. Resources will be adjusted automatically depending on your needs. The nodes will be checked regularly with automatic healing function.
Cloud Storage is an integrated API storage service which can be classified into four categories according to the frequency of access (Frequently access across regions / frequently access within the same region / less than once a month / less than once a year). The data can also be transferred within the four categories as needed. In addition, Cloud storage has the following characteristics: low latency, strong data protection, Object Lifecycle management (setting a Time to Live (TTL) for objects, archiving older versions of objects, or “downgrading” storage classes of objects to help manage costs). You can also access to data through the same set of APIs.
It belongs to one of the relational database services that maintain, manage, and administer your relational PostgreSQL, MySQL, and SQL Server databases in the cloud. It also automate all backup, replication, patching, and update jobs as long as to ensure over 99.95% availability. Each execution unit can provide up to 10 TB of storage, 40,000 IOPS, and 416 GB of memory.
Cloud Bigtable is a high-efficiency, massively scalable service that supports access from all Google Cloud products. It is designed to address a large number of storage types and requests. It is suitable for long-term large-scale analysis and workloads. IT can be scaled to support hundreds of petabytes and handle millions of operations per second. You are able to support higher number of requests per seconds by adding more VM clusters. You don’t need to reboot the machines for configuration changes. All changes will be active immediately without any downtime.
BigQuery is a serverless, highly-scalable, and cost-effective cloud data warehouse with an in-memory BI Engine and AI Platform built in. You can just upload TB or even PB level amount of data to BigQuery. Query by using the ANSI SQL standard language, then Google’s high performance infrastructure can return query results efficiently and quickly. Meanwhile, BigQuery also supports RESTful API, other application can be integrated with BigQuery easily. Since BigQuery is a PaaS by Google, users can just forget how to manage the underlying IT infrastructure and just focus on the tasks.
A.I. / Machine Learning
AutoML includes a user-friendly graphical user interface that allows any users to apply machine learning by applying models that fit to their business needs even the users have only limited machine learning knowledge. Users can build their own machine learning models or choose the models that Google has trained in advance. The trained models cover areas including vision, natural language & structured data. It would be easy for you to select different models based on your business needs.
AI Platform is an integrated platform built by Google. Data engineers and machine learning experts can make use of it to simplify the processes from development phase to deployment phase. The platform provides integrated tools that cover all your needs during the data engineering job with high flexibility and agility. Users can build and run their own machine learning applications based on their own requirements.
Google Cloud’s AI Hub is a hosted repository of plug-and-play AI components. End-to-end AI pipelines and out-of-the-box algorithms are available. Also, users can access the updated AI information provided by Google AI partners.
Sharing function is available so organizations can host their AI content securely. Developers and collaborators within organizations can access information together. It encourages reuse and collaboration of AI contents within the organization.
GCP customers can use the services by just simple operations and application import. Users can focus on application development and management. Moreover, GCP provides a wide variety of industrial solutions. With the consultancy services from Microfusion’s cloud architects, customers can achieve the most suitable solution planning.
Machine Learning is the foundation of AI.
It makes machines to learn how to analyze data or features. Google provides different modules of AI services, including Speech Recognition, Vision Recognition, Language Translation, etc.
It converts your speech to text and is able to recognize 120 different languages. It also supports both real-time audio stream and audio recording as input.
It converts your text to speech and is able to recognize 32 sounds and various languages, which can be applied to different apps.
Dialogflow is an end-to-end, build-once deploy-everywhere development suite for creating conversational interfaces for software, websites and messaging platforms. It incorporates Google machine learning to help create the best online communication ever.
Google Cloud’s Vision API offers powerful pre-trained machine learning models through REST and RPC APIs. Assign labels to images and quickly classify them into millions of predefined categories. Detect objects and faces, read printed and handwritten text, and detect emotions.
Video Intelligence API has pre-trained machine learning models that automatically recognize a vast number of objects, places, and actions in stored and streaming video. Also it can provide insights from video in almosts real time using the Video APIs, and trigger events (content highlights, video recommendations, etc.). It also enhances the captions generation and inappropriate content filtering efficiency.
Google uses machine learning technology to reveal the structure and meaning of texts. You can extract information such as people, places, events, etc. You can also have a better understanding on social media sentiment from customers’ comments and conversations. The Natural Language API also enables you to import analyzed texts to Google Cloud Storage.
By machine learning technology, it is possible to translate texts into more than one hundred languages instantly. You can even apply custom translation models to meet your own business requirements.