Before knowing what exactly Google computing is, let us have a look at what Does basic cloud computing means. Cloud computing provides cloud storage and along with that it also offers computing power. It does not allow users to have direct management. Large clouds provide Various functions Which are located over multiple sources. Its source acts as a data centre for that cloud.
The basic function of cloud computing is to achieve coherence. And the prime challenge that it deals with is providing services to economies of scale.
Google computing runs on the same infrastructure that Google uses for its services. It takes help of Google search, gmail, YouTube and file storage in order to serve its end users. Along with Basic cloud computing services it also offers data analytics, machine learning and data storage. In order to register for Google cloud, customer should provide credit card details or bank account details.
Google computing actually started in April 2008. Google announced its app engine in order to develop and host web applications. Those web applications were managed in data centres of Google. This was the first cloud computing service offered by Google. Afterwards it was scaled and provided for use in November 2011.
Google computing includes Google cloud platform, Google workspace and public cloud infrastructure. In order to handle a huge workload Google set up Google compute engine in June 2012. It was a Pilot project back then but Google made it available to customers in April 2013. Google compute engine is also pronounced as GQ. Every Google compute engine has a disc resource which is known as persistent disk. Persistent disks also act as a raw block device. Persistent disks provide reliable storage at a reliable price.
Customers do not need a separate local ephemeral disc. A single persistent disk has the capacity to be attached to multiple instances. It works in read only mode. Each persistent disk has size up to 10 TB. Google computing encrypt the persistent disks. Encryption is done before data lives the virtual monitor. HMAC scheme maintains the integrity of persistent disk. Google compute engine use the KVM machine. It acts as a hypervisor and supports guest images.
Google compute engine of Google computing offers discounts to its users. When25% of billing cycle is done by one instance, the price starts decreasing. If an instance is used for more than 15 days then user gets 10% discount over the on demand prices. If an instance is used for more than three fourth of the month and the user gets 25% discount over on demand price. If an user uses an instance for an entire month then he or she will get more than 30% discount.
Let us have a look at the resources that Google computing uses in its functioning.
An image resource consist of operating system and route file system. This both are necessary in order to start an instance. Google has a depository of images that are ready to use. The user can also customise an image for Creating an image of choice. If needed user is able to apply an image to persistent disk. The number of cores, memory and I/O operations I determined by type of instance machine.
Persistent disks have a more lifespan than and instance. They are independent virtual machines. The information that is stored here is encrypt it. Encryption is done before it is written to physical media. Each instance has a limited ability to be attached to certain number of total persistent disk space. Usually when instance has the capacity to have 64 TB. Along with that one instance can attach independent persistent disks. For obtaining higher availability, regional persistent disks can be replicated between two zones.
Persistent disks allow users to copy data from existing persistent disks by snapshots. The snapshots can be applied to new persistent disk. Sometimes there is a danger of an unexpected failure. In such situations snapshots can be used for creating backups of persistent disk data.
Google compute engine instance either runs on Linux or Microsoft Windows configuration. It is a virtual machine. Users have the luxury to customise hardware, disc, OS and other configuration options in order to modify the instances.
The address range and get the addresses of all instances connected to it is known as network. Network is the way by which instances communicate with each other, with other networks and with outside world. Each instance is associated with single network. In order to have any kind of communication between the instances present in different networks, it must go through public IP address.
Cloud platform project may contain multiple networks. Multiple instances can be attached to each network the network range for instance is attached to certain network and gateway IP can be defined by user. Usually only one network is needed by users but they get upto five networks per project by default. Each instance is associated with only one network and its network belongs to only one project.
An ephimeral IP address is automatically assigned to instance by default when an instance is created. This IP address stays attached to instance for its entire life. It gets released only when the instance gets terminated. The mechanism to reserve and attach static IPs is to the VMs is provided by GEC.
Each firewall resource is associated with only one network. It is impossible to associate one firewall with more than one networks. No communication is allowed between instances and less firewall resource permits the network traffic. Even when there has to be communication between Instances attached to same network then also firewall permit is required. In order to route the traffic destined for a certain IP range, Google compute engine offers routing table.