Tech Primer: Cloud Technology. Unclouded.

Apart from Big Data that was discussed in the earlier edition, the other crowd-puller in IT world today is Cloud Computing. It offers career opportunities in a range of areas including cloud infrastructure, network, administration and software development.

Lets get down to basics – what is Cloud and why do we need it? Imagine you are a distributor of some goods in a major city. The warehouse space requirement for your stock ranges ranges from as little as 1000 sft to 50000 sft at times. It also overall rises year on year with your company’s growth.

There are several questions before you. How much space to buy ? Where to buy? How much equipment to buy? Will I get more contiguous space whenever I need? How many people to deploy? How to oversee? Sizing for peak loads will result in redundancy during off-peak times, and for less than peak size will result in issues during peak times. City locations are expensive while distant ones mean high transport and logistics costs.

Now imagine if there is a warehouse space provider in your city who will provide space per your dynamically changing requirements; provide exclusive security; will pick-up, package and deliver stocks for you; deploy experts to manage operations; keep it running & accessible 24×7, but finally charge you only for your exact usage! (Subject to a minimum rental of-course.)

Doesn’t this look like a panacea for all your problems? Doesn’t it save a lot of cost, effort and management overhead for you? You get access to only what you want; it is the provider’s headache to manage everything else between customers without impacting each other – be it space, staff, equipment, management or other aspects.

This is the theory for cloud technologies. Your requirements for several needs are taken care of on-demand by the cloud provider, so you don’t need to worry about how much to buy, lead times to buy, where to maintain and how to maintain. It could be software, hardware, utilities or infrastructure in the form of storage space, servers, processing, data centre, hosting, email, desktops, service desk, database etc and so on. Hence it is about optimization and reducing redundancy also.

Web mails(Gmail, Yahoo etc), Facebook and Youtube are some of the easiest examples that everyone can relate to. Even the telephone connection and electricity are a form of cloud delivery!

Service Models

Cloud service models indicate what level of provision or service is needed by the customer. Behind the screens, everything else is managed by the cloud provider. It is generally provided at these 3 levels.

1.  Infrastructure as a Service (IaaS): The fundamental resources like processing, storage and networks is made available by the provider virtually from a shared infrastructure. Customers can visualize as if they have their own complete and closed IT infra or data center on which they can setup their IT environment including servers, OS, network and applications. The Provider will take care of the hardware and aspects like location, scaling, security etc. Billing is usually on the amount of resources allocated and consumed. E.g: Amazon’s EC2, Rackspace.

2. Platform as a Service (PaaS): A computing platform is made available by the provider including the development environment, servers and database. Consumers can create applications using the programming languages and tools provided by the provider and use them commercially. The provider will take care of hardware, servers, network, OS and storage. E.g: Google App Engine, MS Azure, Amazon Beanstalk.

3.  Software as a Service: The providers make readymade software available for usage. Consumers need to know very little about technology and can start using it immediately through internet or secure networks. No installations are needed. Payment is generally on the number of users given access to the application; the number of features enabled; the level of facilities provided like memory, storage, processing etc. Everything underneath about hardware and software, including setup, licenses and maintenance is taken care of by the provider. E.g: Adobe CC, Salesforce.com, Gmail, Yahoo Mail.

Related: Software Testing – A Time Tested Career

Implementation Models

For a customer who chooses to go the Cloud way, implementation can be done in different models:

1. Private Cloud: When cloud infra is deployed exclusively for a single organisation, hosted mostly internally within the organisations premises or sometimes externally at the provider’s data center. This model is capital intensive and does not implement quite a few of the optimization aspects of cloud. The driving factors for this model are generally regulatory and security considerations. E.g Banks and financial institutions in many countries are not allowed to take data storage outside the bank’s premises or into shared environments.

2. Public Cloud: The provider hosts the environment from its own data center and access is provided over a separate network or public internet. Anyone can pay and securely engage the services. This is the most common and economically beneficial implementation of cloud that utilizes its core concepts. E.g. Amazon AWS, Microsoft and Google

3. Hybrid Cloud: It is a combination of private, public and other models like community cloud services, from different providers. A hybrid cloud service crosses isolation and provider boundaries so it can’t be simply put in one category of private, public, or community cloud service. The uses of this model are very specific and complex, like for a global organisation which has secure as wells as public servers and applications on cloud, customized to regional requirements and regulations.

The term Virtualization is synonymous with Cloud. Virtualization allows creating multiple virtual systems (Server, Network, Storage, Desktop, Application or Mobile) from a common set of physical infrastructure.

Each of the virtual systems is segregated from the other logically. For e.g. if you want to test a piece of software on 10 different OSs, you can create 10 Virtual desktops on a single desktop, each with different OS instead of setting in 10 different physical computers. Each instance appears like an independent desktop. The life of the Virtual desktop is limited to the session.

In a cloud implementation, each customer or customer organisation sees only the virtual system (Infra, Platform or Software that is provided as a Service) provided to it, which is as per its required specifications like OS, processing, network, storage, memory, application etc resources. In the background the cloud provider uses a large common infrastructure that they scale-up or scale-out with growth.

Software called a Hypervisor enables virtualization and is used for generation and management of Virtual instances. Certain types of Hypervisors run directly on Hardware and some on an OS. E.g. Citrix XenServer, KVM, VMware ESX/ESXi, Microsoft Hyper-V.

Key benefits for Organisations moving to cloud

– Huge savings on infrastructure, manpower, space, power, time-to-market.

– Need not worry about provisioning for expansion requirements, peak loads or redundancy for long periods of lows.

– Applications and systems can be rolled out at very short notice to staff or customers across the world.

– Need not worry about hardware and software licences and theiradvance renewals.

– Can ensure their systems are always up-to-date with the latest versions of software and applications.

– Need not worry about maintenance of infrastructure, support, replication, backup and disaster recovery.

Related: A Beginners Guide to Big Data

Concerns and Risks

The key concerns and risks about Cloud are around the safety, security and privacy.

The cloud provider hosts data and applications of many competing customers and its admin staff may have access to private data of each its customers.

This gives rise to worries around potentially illegal access to confidential data by a competitor, accidental deletion or loss of data, hacking of cloud servers, service downtime, quality of service etc.

While such concerns exist, most cloud providers offer a very high level of security and are investing billions into making Cloud more robust and secure.

Opportunities

Cloud Computing business provides technical career options in several areas like architecture, system design, hardware and software provisioning, deployment, cloud administration, server and network management, remote support, virtualization, service management, security, database management, backup, recovery, replication, and cloud specific software and tools.

All these are interlinked. Hence people wanting to switch career to Cloud related areas usually need to have a combination of knowledge and skills in some of these aspects.  The nature of work is very different from regular application development and support.

With increasing awareness about cloud and organisations trying to harness its benefits, opportunities to work in these areas are ever increasing. If you are in for the challenge, go for it. www.addmylearning.com is there for you to provide support in getting the best Cloud training options from different trainers.