Ten promising ways to leverage the cloud-first era2019-02-01T11:36:48.000Z 2019-02-01T11:36:48.000Z Although it has been a while since the concept of cloud computing first appeared, it seems that the time of its broad applicability has come only now. Serverless computing gained its popularity in 2014 thanks to the creation of AWS Lambda. A bit later the release of several other products such as Google Cloud Functions, IBM OpenWhisk on Bluemix accompanied by the introduction of other serverless frameworks comprising Fission, Fn, Terraform has strung along with AWS Lambda.
Although it has been a while since the concept of cloud computing first appeared, it seems that the time of its broad applicability has come only now. Serverless computing gained its popularity in 2014 thanks to the creation of AWS Lambda. A bit later the release of several other products such as Google Cloud Functions, IBM OpenWhisk on Bluemix accompanied by the introduction of other serverless frameworks comprising Fission, Fn, Terraform has strung along with AWS Lambda.
According to common definitions, serverless computing is the idea of development and release of the applications which do not need server management. Putting this another way, it means the hyper-focused production model based on which the applications, combined into single or multiple functions, are placed to the platform, then executed, upscaled and expensed further to a specific requirement existing at that point of time. The advantages of serverless computing comprise zero server ops and the possibility to save computing resources in the off-line state.
In fact, the term “serverless” doesn’t mean that the servers are not used at all. The servers are still deployed for hosting and running the code. At the same time, serverless computing has eliminated the requirement for server navigation, maintenance, upgrading, upscaling and planning of capacity. Now all these functions are performed by a serverless ecosystem and are isolated from the development and operation teams. Therefore, the developers can concentrate on coding the business logic layers. It is against this background that the operating personnel can give a boost to more crucial tasks.
The most applicable areas for serverless computing comprise the processes which are asynchronous, parallel, rare, occasional, incalculable fluctuations in upscaling requirements, structureless, transitory and fast-moving. In a nutshell, serverless technology will not be the best choice for the users who strongly depend on time and system capacity of the startup or for those who don’t want to be tied with an exact cloud base provider.
The Cloud Native Computing Foundation determines ten most promising areas for serverless computing application:
Processing of multimedia. It provides an introduction of activities that perform conversion operation as a reaction to uploading the file;
Changing databases or collecting change information. Inspection of the changes to verify their compliance with the quality criterion;
Requests being entered into the IoT. The function allowing answering the inquiries and upscaling in reply;
Processing the upscaled stream: handling data in the unlimited stream of requests;
Chatbots: automatic upscaling in case of extreme inquiries;
Bundle tasks. The set of tasks that has to be performed simultaneously and might require IO and access to the network;
Web applications and HTTP REST APIs: common inquiries and respond operating load;
Shifting back end: development on the REST API backend operating load on the top of BaaS APIs;
Domain layer: navigation of microservice operation loads that performs a range of stages;
Sustained integration tunnel: the function of eliminating the necessity for preliminary supplied hosts.
According to the CNCF, serverless computing is the next stage of cloud-first coding. Keeping its eye on the big picture, the CNCF expects to propel integration of serverless technology via cooperative strivings accentuating the capacity for interaction. A year ago, the CNCF has established the Serverless Working Group comprising the range of eminent companies such as IBM, Microsoft, Google, Oracle etc.
Lately, the CNCF has introduced the open source specification CloudEvents to the Sandbox. Therewith CloudEvents strives to create the specifications for each type of event data. Since the developers are lacking a standard method to describe the events, they have to continuously practice obtaining the events. Besides, it restricts the prospects for libraries, instruments, and substructures to support supplying event data across the ecosystems such as System Development Kits, tracking systems and so on. The mobility and efficiency that can be obtained from event data are obstructed.
This startup will assist the CNCF in the attempts to advance serverless computing technology and to allow the developers launching and controlling the application code including the actions started by the events.
One conclusion can be definitely drawn: being a new tech, it is experiencing a shortage of unification and reciprocity between cloud providers which may result in isolation of a vendor. Quality instructions, approaches and, essentially, instruments and ecosystems are required now. It is crucial to interconnect multiple operators on the same platform to boost innovative technology via tight cooperation.