Cloud Computing vs. Grid Computing

The terms “cloud computing” and “grid computing” both suggest data centers and computing resources that are available over networks, leading many to believe that the two are interchangeable. This article looks at the fundamentals of cloud and grid computing, the major differences between the two and the advantages and disadvantages of each.

Grid Computing: the Basics

Grid computing links disparate computers, forming a single unified infrastructure. This facilitates the provisioning of computing resources, much like a public utility that can be “on” or “off.” Grid computing relies on software (middleware) that divides and directs pieces of a program as one large image to a number of different computers.

For these reasons, it is common to describe grid computing using an electrical grid analogy. In an electrical grid, wall outlets link users to an infrastructure of resources that generate, distribute and bill for electricity. With grid computing, middleware coordinates varied IT resources across a network. Grid computing links the disparate parts as a virtual whole. The objective of grid computing is to give users access to IT resources when they need them.

Grid Computing: the Benefits

Grid computing allows organizations to meet two goals:

1. Remote access to IT assets
2. Aggregated processing power

Grids are made up of processors, sensors, data-storage systems, applications and other IT resources, all of which are shared across the network. Grid computing systems have made it possible to carry out a number of resource-demanding projects that had previously been impossible or highly challenging, due to the physical location of vital IT assets.

Cloud Computing: the Basics

Cloud computing is a development on grid computing. It requires three fundamental components:

1. Thin clients (alternatively, clients with a thin-thick switch may be used)
2. Grid computing
3. Utility computing (i.e. paying for resources used from shared servers, similar to paying for a public utility, such as electricity)

While grid computing facilitates access to computing resources as if they were public utilities, cloud computing offers on-demand resource provisioning. This eliminates the possibility of over-provisioning.

The term “cloud computing” is new, in terms of search engine trends. The term first appeared in trend analyses in 2007, surpassed the term “grid computing” in 2008 and continues to gain momentum.

Cloud Computing: the Benefits

Cloud computing provides organizations with massive scalability possibilities, without having to invest resources into developing new infrastructure, training employees or licensing new software. This may be particularly attractive for small- and medium-sized businesses, which may be able to outsource their data-center infrastructure. Larger organizations may also choose to access peak load capacity, without the heavy investment of building internal data centers.

Organizations relying on cloud computing do not own the software, platform or infrastructure. Needless to say, this significantly reduces the upfront costs, operating expenses and capital expenses. Running operations from the cloud eliminates the involvement in server or network maintenance. Finally, users benefit by being able to access multiple servers, from anywhere in the world.

Another incentive for organizations moving into the cloud is the issue of environmental impact. Proponents of cloud computing point out that consolidation of IT assets in a data center may be a more environmentally friendly method of computing, as it reduces the number of hardware components necessary, thus decreasing the consumption of energy for running and cooling the hardware. As cloud computing facilitates telecommunication, it could eliminate the need for office space and commuting, both positive steps in reducing our carbon footprints.

Similarities

As cloud computing represents an advancement from grid computing, there are naturally a number of similarities. These are discussed below:

• Scalability – Scalability refers to the system’s ability to handle increasing amounts of work, or to improve performance. Both grid and cloud computing are scalable, as application instances are load balanced. CPU and network bandwidth is directed on demand. This means that grid and cloud systems have a storage capacity that fluctuates, depending on the number of users, instances and the amount of data transferred at a specific time.

• Multitenancy and Multitasking – Multitenancy refers to a situation in which a single instance of software is able to serve multiple clients (tenants). Multitasking refers to when multiple tasks (processes) share processing resources. Multitenancy and multitasking, common to both grid and cloud computing, allow numerous users to perform different tasks and access single or multiple application instances.

Cloud & Grid: the Concerns

There are a number of potential disadvantages to cloud and grid computing methods. These are briefly outlined below:

• Interoperability – In order to ensure interoperability, it is necessary for decision-makers to examine how different cloud service vendors import or export data. Even though industry cloud computing standards do not exist for APIs (application programming interfaces) or importing/exporting data, a number of vendors are collaborating to ensure interoperability for their clients.
• Hidden Costs – Such costs may include higher network charges for storage and database applications, or latency issues for users who may be located far from cloud service providers.
• Surprises – Experts recommend testing and pilot programs to ensure there will not be any unexpected outcomes. These may include tests checking application validation, allocating/releasing resources, etc.
• Threshold Policy – It is important to consider how the cloud service provider will handle sudden increases or decreases in demand. How will unused resources be allocated?

Summary

This article explores cloud computing and grid computing. Grid computing appeared in the late 1990s to describe technologies that allow users to access computing power on demand. This type of computing is similar in form and utility to an electrical grid. Cloud computing is based on grid computing, and allows users to access shared servers, which distribute resources, software and data on demand. The article looks at the benefits and disadvantages of grid and cloud computing, as well as their similarities and differences.

CCSK Exam Preparation

In preparation for the Certificate of Cloud Security Knowledge (CCSK), a security professional should be comfortable with topics related to this post, including:

• Essential Characteristics of Cloud Computing (Domain 1)
• Multi-Tenancy (Domain 1)
• Resource Sharing (Domain 8)
• Provisioning (Domain 12)