evolution in grid computing

GRID COMPUTING

CONTENTS:

1. ABSTRACT
2. INTRODUCTION TO COMPUTING
3. TYPES OF GRID COMPUTING
4. ISSUES
5. APPLICATIONS
6. A SIMPLE HELLO WORLD EXAMPLE
7. DOWNSIDESOF GRID COMPUTING
8. CONCLUSION

1: ABSTRACT:

Grid computing is said to be the next big thing in IT. Research in grid computing is making rapid progress, owing to the increasing need for large-scale computation in the resolution of complex problems.
Clusters are, in a sense, the predecessors of grid technology. Clusters interconnect nodes through a local high-speed network, using commodity hardware, with the aim of reducing the costs of such infrastructures.
Supercomputers have been replaced by clusters of workstations in a large number of research projects.
Grids provide access to widely distributed computing and data resources, allowing data-intensive applications significantly improved data access, management and analysis. Nowadays there are a huge number of data intensive applications in several domains such as physics, climate modeling, biology/bio-informatics, addressing some of the most important current problems.
Computing grids are conceptually not unlike electrical grids. When you connect to the electrical grid, you don’t need to know where the power plant is or how the current gets to you. Grid computing uses middleware to coordinate disparate IT resources across a network, allowing them to function as a virtual whole.
Grids use a layer of middleware to communicate with and manipulate heterogeneous hardware and data sets. In some fields— astronomy, for example—hardware cannot reasonably be moved and is prohibitively expensive to replicate on other sites.

Grids address two distinct but related goals: providing remote access to IT assets, and aggregating processing power. The most obvious resource included in a grid is a processor, but grids also encompass sensors, data-storage systems, applications, and other resources.
Grids use a layer of middleware to communicate with and manipulate heterogeneous hardware and data sets. In some fields— astronomy, for example—hardware cannot reasonably be moved and is prohibitively expensive to replicate on other sites.
Many grids are appearing in the sciences, in fields such as chemistry, physics, and genetics, and cryptologists and mathematicians have also begun working with grid computing. Grid technology has the potential to significant impact other areas of study with heavy computational requirements, such as urban planning. Another important area for the technology is animation, which requires massive amounts of computational power and is a common tool in a growing number of disciplines included in a grid is a processor, but grids also encompass sensors, data-storage systems, applications, and other resources.

2: INTRODUCTION
TO COMPUTING:

Grid computing is an implementation in an enterprise computing taxonomy .It consists of family of technologies for opportunistically providing computing power from a pool of resources. Grid computing is opportunistic since it has to wait for resources to become available. The resources may include computing cycles, file and data storage, caching, network bandwidth, databases and application software. These resources can be distributed diversely on the globe .Provision of computer power means the methods and mechanisms for locating, authorizing, assembling, scheduling, releasing and accounting for resources and their usage.
For example compute grids share computational and data resources. And some grids share both and can also share network bandwidth, storage and caching resources and application software.

So the first dimension reflects the types of resources that the grid can use .Second dimension describes a grids geographic or administrative reach .Third dimension tells how the companies can get the resources. Fourth dimension reflects the partnerships between the Enterprises. The Fifth dimension reflects for what type of application a grid is used for biology, sensors or any other access.
So this proves that a Grid is multidimensional.
Coming to the Cluster, they are connected commonly connected to a high speed LANS. Clusters are usually deployed to improve speed and/or reliability over that provided by a single computer, while typically being much more cost-effective than single computers of comparable speed or reliability. They can be geographically distributed, but are often closely coupled in the same room.

Clusters can consist of heterogeneous processors and peripherals but they are homogeneous and use high performance and special purpose interconnection networks.


Fig: clusters in a university
But a CLUSTER is not a grid.
Since Computer Clusters require a much higher degree of centralized control, this point clearly distinguishes between a cluster and a grid.
Mainly the grid community focuses on issues such as Reliability, Security, Service and Quality Performance and Resource integration.
3:TYPES OF GRID COMPUTING:

1: Utility Computing:
Here, the main idea is to offer computing resources as an on-demand service to customers in much the same way that utilities offer electrical, gas, water, and telephone services to households and businesses. The utility computing service provider offers hosted computing resources.
One distinction is that the on-demand computing resources can comprise a grid in the service provider’s realm, and the grid can span several sites in the provider’s service area. Grid economies and scalability add a new dimension.





2: Autonomic computing:
Autonomic computing architectures monitor utilization and performance, tune and manage themselves, and adapt to failure. Some share resources and schedule tasks with other systems. The fundamental components of an autonomic computing system provide functions that computing grids will almost certainly need to operate effectively, such as the ability to recover lost computational subtasks.
As grids evolve, they might take on many of the characteristics of an autonomic computing system: Self-monitoring, diagnosis, and adaptability in their youth; sophisticated resource scheduling and forecasting; and perhaps a vertebrate-like involuntary autonomic nervous system at maturity. So the real issue is the extent to which a grid has adopted the characteristics of autonomic computing.




3: Peer-to-peer computing:

Peer-to-peer computing is one type of application that uses grid services to advertise, find, and share files.
The grid community tends to focus on top-down issues such as
1: resource integration,
2: performance,
3: reliability,
4: service quality, and
5: security.
The peer-to-peer community tends to focus on bottom up issues such as narrowly defined and specialized services, and support for tens of thousands of concurrent participants.

4: ISSUES:

The success of grid computing depends on fundamental issues in 2 main areas:
1: security
2: performance

Security:
Grids must deal with every security that any enterprise-owned or outsourced computing model faces. Security issues include secure authentication, access rights and privileges. Reliable and secure communications, perhaps with encryption, are also a requirement. Maintaining confidentiality and privacy will also be issues if you are transferring personal data.

Performance:
For a grid performance is the main key is to deliver nontrivial qualities of service “. Some grid services might fall short because

the scattering and gathering steps can incur significant delay. Grid computing tends to be opportunistic – it must wait for computing resources to become idle – which means that performance can be nondeterministic. Grid performances include resource availability and reliability, utilization and load, response time, delay and delay variation. Data Integrity is another consideration .
5: Applications:

1: Bio-Informatics:
Bioinformatics analysis of data produced by complete genome sequencing projects is one of the major challenges. Integrating up-to-date databanks and relevant algorithms is a clear requirement of such an analysis. Grid computing would be a viable solution to distribute algorithms and data, computing and storage resources for Genomics.
When bioinformatics grid server receives the computational requests from the client, it locates a suitable node in the grid to perform the mathematical computation according to the users’ requirement and task allocation rule, or integrates a virtual supercomputer to perform the larger computational requests from users.
2: SMALLPOX project:
We intend to use grid computing to screen millions of potential anti-smallpox drugs against this target.
The Smallpox Research Grid uses a SETI-like model to analyze interactions between virus protein targets and a catalog of tens of millions of drug molecules. The Smallpox project can harness millions of computers belonging to people in over two hundred countries, all of whom will benefit from protection against smallpox.
By adding your CPU to the global grid, every time your computer is idle, you contribute your computing resources to the grid, accelerating the screening process while dramatically reducing the cost of the project.
The result is that rather than spending years to screen hundreds of thousands of molecules, it will be possible to screen hundreds of millions of molecules in just months.
This saves a large amount of system time and alters the use of Resources.

3: Grid Computing In SETI:
SETI, the search for Extraterrestrial Intelligence uses a huge number of Internet-connected computers to download the results during idle times.
As of late 2004, SETI had scavenged 1.83 million years of CPU time from 4.9 million users in 226 countries .It had used this grid to perform 4.5*10( power)21 floating-point operations Experts say that 10(power)21 -one sextillion-is the approximate number of grains of sand on all of Earth's beaches and deserts
It is just one order of magnitude shy of the estimated number of stars in the visible universe.



Some SETI observations have been conducted using the radio telescope

4: The Rice Genome Program: :
Bioinformatics research leads to a lot of information collected worldwide for which large databases should be maintained .Grid computing solves this problem. Grid computing would be a viable solution to distribute algorithms and data, computing and storage resources for Genomics.
Finding a single genome of Rice (scientific name Oryza sativa) can take a lot of time which measures up to months examining all the base pairs which are in millions These data is stored in data repositories. Hence we require a technology not only to visualize analyze DNA data but also the integration and exchange of information on a gene or coding regions from different international collaborative databases needs to be done in a careful and in a robust manner .Grid Technology Solves this problem.

Grid Technology enables sharing of bioinformatics data from different files by creating a virtual organization of data.
So the collection of databases generated during the research work on RICE is collected and maintained by a grid which helps in posing Queries on a particular type of Rice.


Here the user must provide the services they want providing constraints like family size , protein interactions ..etc .


6: A simple Hello world Example:

Let us consider some cases of web client-server applications like Video streaming Game serving, File downloading etc. So an approach to provide a scalable solution is to distribute the application even to the other servers that run the same application. A Network dispatcher is the entry point for an application but does not run the application , rather the servers which are connected onto the same LAN handle the workload and answers the queries of the client.
Let us consider an example: A network dispatcher (or) Front end server waits for client requests. When connected the client is given back a ticket and an application service IP address of where to connect.
The application answers HELLO WORLD when the client connects to it . The application is started on the application servers by the front end server.
The Executable for the client is Hello client and takes the front end server host name as parameter.


Grid computing used in Hello world Example .
7:Downsides of Grid Computing:

Being able to access distant IT assets—and have them function seamlessly with tools on different platforms—can be a boon to researchers, but it presents real security concerns to organizations responsible for those resources. An institution that makes its IT assets available to researchers or students on other campuses and in other countries must be confident that its involvement does not expose those assets to unnecessary risks. Similarly, directors of research projects will be reluctant to take advantage of the opportunities of a grid without assurances that the integrity of the project, its data, and its participants will be protected.
Another challenge facing grids is the complexity in building middleware structures that can knit together collections of resources to work as a unit across network connections that often span oceans and continents. Scheduling the availability of IT resources connected to a grid can also present new challenges to organizations that manage those resources. Increasing standardization of protocols addresses some of the difficulty in creating smoothly functioning grids, but, by their nature, grids that can provide unprecedented access to facilities and tools involve a high level of complexity.
A word of caution should be given to the overly enthusiastic. The grid is not a silver bullet that can take any application and run it a 1000 times faster without the need for buying any more machines or software. Not every application is suitable or enabled for running on a grid.


Some kinds of applications simply cannot be parallelized. For others, it can take a large amount of work to modify them to achieve faster throughput. The configuration of a grid can greatly affect the performance, reliability, and security of an organization’s computing infrastructure. For all of these reasons, it is important for us to understand how far the grid has evolved today and which features are coming tomorrow or in the distant future.

8: CONCLUSION:

A greater awareness of this area is needed so that people can make a direct contribution towards solving these problems by providing whatever spare computing resources they may have at their disposal (usually in the form of idle cycles). Grid computing is the future for Bio-Informatics which helps to create an epidemic free future for man kind and to solve long-term problems. Grids make research projects possible that formerly were impractical or unfeasible due to the physical location of vital resources.
Using a grid, researchers in Great Britain, for example, can conduct research that relies on databases across Europe, instrumentation in Japan, and computational power in the United States. Making resources available in this way exposes students to the tools of the profession, facilitating new possibilities for research and instruction, particularly at the undergraduate level.
Although speeds and capacities of processors continue to increase, resource-intensive applications are proliferating as well. At many institutions, certain campus users face ongoing shortages of computational power, even as large numbers of computers are underused. With grids, programs previously hindered by constraints on computing power become possible.