BASICS OF GRID AND CLOUD COMPUTING

Basics of Grid Computing

Grid Computing harnesses distributed resources from various institutions (resource providers), to meet the demands of clients consuming them. Resources from different providers are likely to be diverse and heterogeneous in their functions (computing, storage, software, etc.), hardware architectures (Intel x86, IBM PowerPC, etc.), and usage policies set by owning institutions.

Developed under the umbrella of Grid Computing, information services, name services, and resource brokering services are important technologies responsible for the aggregation of resource information and availability, selection of resources to meet the clients’ specific requirements and the quality of services criteria while adhering to the resource usage policies.

Figure 8.1 shows an exemplary relationship of resource providers and consumers for a collaborative Grid computing scenario. Clients or users submit their requests for application execution along with resource requirements from their home domains.

A Resource broker selects a domain with appropriate resources to acquire from and to execute the application or route the application to domain for execution with results and status returning to the home domain.

Basics of Cloud Computing

IDC1 defined two specific aspects of Clouds: Cloud Services and Cloud Computing. Cloud Services are “consumer and business products, services and solutions that are delivered and consumed in real-time over the Internet” while Cloud Computing is “an emerging IT development, deployment and delivery model, enabling real-time delivery of products, services and solutions over the Internet (i.e., enabling Cloud services)”.

Amazon’s Elastic Compute Cloud2 popularized the Cloud computing model by providing an on-demand provisioning of virtualized computational resources as metered services to clients or users. While not restricted, most of the clients are individual users that acquire necessary resources for their own usage through EC2’s APIs without cross organization agreements or contracts.

Figure 8.2 illustrates possible usage models from clients C1 and C2 for resources/services of Cloud providers. As Cloud models evolve, many are developing the hybrid Cloud model in which enterprise resource brokers may acquire additional needed resources from external Cloud providers to meet the demands of submitted enterprise workloads (E1) and client work requests (E2). Moreover, the enterprise resource domain and Cloud providers may all belong to one corporation and thus form a private Cloud model.

DISTRIBUTED COMPUTING IN THE GRID AND CLOUD BASICS

The Grid encompasses two areas of distributed system activity. One is operational with an objective of administrating and managing an interoperable collection of distributed compute resource clusters on which to execute client jobs, typically scientific/ HPC applications.

The procedures and protocols required to support clients from complex services built on distributed components that handle job submission, security, machine provisioning, and data staging. The Cloud has similar operational requirements for supporting complex services to provide clients with services on different levels of support such application, platform and infrastructure.

The Grid also represents as a coherent entity a collection of compute resources that may be under different administrative domains, such as universities, but inter-operate transparently to form virtual organizations.

Although interoperability is not a near term priority, there is a precedent for commercial Clouds to move in this direction similarly to how utilities such as power or communication contract with their competitors to provide overflow capacity.

The second aspect of distributed computing in the Grid is that job themselves are distributed, typically running on tightly coupled nodes within a cluster and leveraging middle ware services such as MPICH. Jobs running in the Grid are not typically interactive, and some may be part of more complex services such as e-science work flows.

Workloads in Clouds usually consist of more loosely coupled distributed jobs such as map/reduce, and HPC jobs written to minimize internode communication and leverage concurrency provided by large multi-core nodes.

Service instances that form components of a larger business process work flow are likely to be deployed in the Cloud. These workload aspects of jobs running in the Cloud or Grid have implications for structuring the services that administer and manage the quality of their execution.

CLOUD COMPUTING VENDOR LANDSCAPE BASIC INFORMATION AND TUTORIALS

At the beginning of the new millennium there was not yet such a thing as a cloud computing vendor, though (as we have seen) teams were already hard at work on several significant efforts . . . and several of these eventually blossomed into key cloud computing vendors.

In fact, only a few years after a modest beginning for the nascent industry, there is a vibrant vendor ecosystem, with everything from relatively established players to the latest, most hopeful startups, and much in between. Hardly a month passes without numerous, significant  product announcements, nor a quarter without new vendors and open source projects.

Cloud computing is clearly an area of rapid evolution. As a result, in order to ensure the most useful (current) information this brief appendix contains information that is least likely to change rapidly—overview information for the major categories, including examples of some of the vendors in each category.

Comprehensive, current listings of companies and products, including industry trends and recent developments are available on the web site. The major categories include three that correspond to the major layers of the cloud technology stack, and two for those providing expertise in one form or another. Each category includes vendors focused on public, private, and hybrid cloud offerings; those focused on commercial as well as government markets; startups and the established; open source, open distribution, and traditional distribution models; and in many cases, all of the above.

Of course certain vendors have offerings in more than one category; a handful intend to cover each category, though that will likely be difficult to achieve and maintain. In any case, here are the major categories, along with a few notes about the history that shaped each category.

Infrastructure as a Service (IaaS)
Vendors in the Infrastructure as a Service (IaaS) category primarily fall into two broad groups: those that provide an existing IaaS and those that provide technology to enable IaaS. Vendors that provide an existing IaaS generally come from cloud technology providers (e.g., Amazon), managed services or hosting providers (e.g., Rackspace, Savvis, etc.), and integrated vendors such as HP, IBM, and Dell.

The technology providers include those who provide software stacks to manage physical or virtualized infrastructure (such as VMWare) as well as those who provide hardware (of varying degrees of commodity) that is intended for easy stacking, replacement, and so forth (all of the major hardware providers, several startups, and certain fresh entrants from nontraditional vendors, such as Cisco).

This is a category that is likely to see significant innovation– in particular, as the trend towards commoditization of the infrastructure matures, then very-high volume providers of commodity infrastructure are likely to dominate, both amongst the ready to consume IaaS and the technology providers.

CLOUD COMPUTING PLANNING STAGE TACTICS BASIC INFORMATION

At the phase of cloud planning, it is necessary to make a detailed investigation on customer position and to analyze the problems and risks in cloud application both at present and in the future. After that, concrete approaches and plans can be drawn to ensure that customers can use cloud computing successfully to reach their business goals.

This phase includes some practicable planning steps in multiple orders listed as follows,

(1) Business Architecture Development
While capturing the organizational structures of enterprises, the business models also get the information on business process support.

As various business processes and relative networks in enterprise architecture are being set down one after another, gains and losses brought by relative paths in the business development process will also come into people’s understanding.

We categorize these to business interests and possible risks brought by cloud computing application from a business perspective.

(2) IT Architecture Development
It is necessary to identify the major applications needed to support enterprises business processes and the key technologies needed to support enterprise applications and data systems. Besides, cloud computing maturity models should be introduced and the analysis of technological reference models should be made, so as to provide help, advices and strategy guide for the design and realization of cloud computing mode in the enterprise architecture.

(3) Requirements on Quality of Service Development
Compared with other computing modes, the most distinguishing feature of cloud computing mode is that the requirements on quality of service (also called non-functional needs) should be rigorously defined beforehand, for example, the performance, reliability, security, disaster recovery, etc.

This requirement is a key factor in deciding whether a cloud computing mode application is successful or not and whether the business goal is reached; it is also an important  standard in measuring the quality of cloud computing service or the competence in establishing a cloud computing center.

(4) Transformation Plan Development
It is necessary to formulate all kinds of plans needed in the transformation from current business systems to the cloud computing modes, including the general steps, scheduling, quality guarantee, etc. Usually, an infrastructure service cloud cover different items such as infrastructure consolidation plan report, operation and maintenance management system plan, management process plan, application system transformation plan, etc.

CLOUD COMPUTING SECURITY BASIC INFORMATION

One of the biggest user concerns about Cloud Computing is its security, as naturally with any emerging Internet technology. In the enterprise data centers and Internet Data Centers (IDC), service providers offer racks and networks only, and the remaining devices have to be prepared by users themselves, including servers, firewalls, software, storage devices etc.

While a complex task for the end user, he does have a clear overview of the architecture and the system, thus placing the design of data security under his control. Some users use physical isolation (such as iron cages) to protect their servers. Under cloud computing, the backend resource and management architecture of the service is invisible for users (and thus the word

“Cloud” to describe an entity far removed from our physical reach). Without physical control and access, the users would naturally question the security of the system.

A comparable analogy to data security in a Cloud is in financial institutions where a customer deposits his cash bills into an account with a bank and thus no longer have a physical asset in his possession. He will rely on the technology and financial integrity of the bank to protect his now virtual asset.

Similarly we’ll expect to see a progression in the acceptance of placing data in physical locations out of our reach but with a trusted provider. To establish that trust with the end users of Cloud, the architects of Cloud computing solutions do indeed designed rationally to protect data security among end users, and between end users and service providers.

From the point of view of the technology, the security of user data can be reflected in the following rules of implementation:

1. The privacy of user storage data. User storage data cannot be viewed or changed by other people (including the operator).

2. The user data privacy at runtime. User data cannot be viewed or changed by other people at runtime (loaded to system memory).

3. The privacy when transferring user data through network. It includes the security of transferring data in cloud computing center intranet and internet. It cannot be viewed or changed by other people.

4. Authentication and authorization needed for users to access their data. Users can access their data through the right way and can authorize other users to access.

CLOUD COMPUTING STRATEGIC BUSINESS AND FINANCIAL IMPLICATIONS

The challenging economy made the cloud computing conversation especially relevant. The business and financial potential of cloud makes it a special trend for us to embrace. We will delve deeper into the full range of business and financial benefits later. The strategic business and financial implications of cloud are the focus of this article.

First and foremost,  with cloud computing, we have another avenue for realizing business agility, the Holy Grail of all business strategies. As with all technology trends, business agility is probably the most frequently mentioned goal of business and technology executives when they describe their strategies, and yet it remains the least realized in terms of execution.

We could even go so far as to say that a clearly articulated business or technology strategy that can deliver on that promise, that is clearly articulated, and has been incorporated into daily operations can seem as elusive as any mythological beast. Fortunately, this opportunity truly is different.

Cloud computing offers business agility in a simple, clearly understandable model: For a new startup or for emergent business requirements of established enterprises, cloud computing allows an organization to implement a rapid time-to-market model by securely accessing a ready-to-use IT infrastructure environment, hosted and managed by a trusted third party, with right-sized, scalable computing, network and storage capability, that we pay for only as we use it and based on how much of it we use. Hmmm, let me think about this a while . . . NOT!!!

We do not have to build or expand our data center (no construction of buildings, raised floor, energy and cooling equipment, building automation and monitoring equipment, and no staff); we do not have to buy any hardware, software, or network infrastructure (no dealing with the procurement hassles we are so accustomed to, especially with the inevitable delays in IT acquisition); we can rapidly implement a new business model or start a new company to address a new market need far faster than we normally could have; and we do not have to continue to pay for the cloud infrastructure and resources if we discontinue the project or if the company fails.

From a business and IT executive’s perspective, what is not to like about this business vignette?

There are countless new startup firms that have leveraged cloud computing models to obtain their IT infrastructure as a service, therefore enabling them to focus their limited funds and resource bandwidth on their unique technology and business model innovation.

Resource constraints are liberating in this sense, since they force new startups to leverage ready-to-use cloud resources as opposed to building a data center. These types of scenarios, of course, raise a number of business and financial implications that must be explored further.