ISCA Ch 8 (Part 1)

Chapter 8 
Emerging Technologies
PART 1


                                                                   Cloud Computing

What is Cloud Computing?

As the name suggest, it is Computing done through internet. Its a combination of SOFTWARE, HARDWARE AND NETWORK. It provides anywhere access.

Imp. pts: 
1.)  Location of physical resources may not be known to user.
2.)  Pay as you use. 

It provides simple front end interface like web browser.

For Eg. Google Apps

Now two terms will be used in this chapter. DYNAMIC VS. SCALABLE.

Dynamic: Which can be changed as per need.
Scalable: Which can be expanded as per need.

For eg;
Most Cloud Service providers will allow you to increase your existing resources to accommodate increased business needs or Changes. This will allow you to support your business growth without expensive changes to your existing IT Systems.

                                           
                                            CLOUD COMPUTING VS GRID COMPUTING

First of all we need to understand grid computing.

In Grid Computing we divide a task between thousands of computers (Computers are connected to each other). So as a result, a task is performed by thousand of computers and in the end we accumulate the result of all computers. So it can be used only for large problems only.

Biggest Flaw: If processing on one computer fails the whole program will not work.
Solution: If we use a failover Component. ( i.e. allot the work given to failed computer to another computer.)

Now lets understand the difference between Cloud and Grid Computing.

Cloud Computing evolved from Grid Computing.

Similarity:
1.) Scalable Services by using Load Balancing (i.e. distributing workload on different machines as per the available capacity)

2.) Multitasking: It means many customers can perform different tasks at the same time.

3.) Both Computing provides service level agreements for guaranteed uptime availability.

Differences:
1.) Grid can be used only for large data storage while cloud can be used both for small and large data storage.

2.) Gird can be used for Large and complicated problems only while cloud can used for both large and small computations.

3.) Under grid there are more chances of failure while under cloud there are less chances of failure. (remember failover component.)

PERTINENT ISSUES

1.) Threshold Policy: It is a Policy for allocation of resources to clients as per their work load.
2.) Interoperability: Suppose we have two cloud vendors. Now both use their different API (Application Program Interface), different forms for importing and exporting data. Now if MR. A avail services of 1st vendor and later on wants to change to 2nd vendor. Here comes the problem of interoperability because of different forms, API.

3.) Hidden Cost: There are Various hidden cost. These are High Data Charges, High Storage cost, Latency Cost.

(Latency: In Computer Network, it is an expression of how much time it takes for a packet of data to get from one designated point to another.)

Unexpected Behavior: Cloud Applications may not work properly in Cloud. i.e they are working fine on a single computer but as we launch the application in cloud for sharing the app on other computers it shows unexpected behavior (i.e. error). NOTE: There is no problem in application programming but problem is in cloud server itself.

4.) Security Issues: There are many security issues.
      * Recovery of Data from cloud.
      * Time Taken in recovery of data.
      * Theft of Data.

Solutions to above problems:

     * Contract with Could service provider. ( i.e service credits for loss of data.)
     * Checksum match with data recovered to check whether data downloaded is same as uploaded.
     * Safe keeping of your Private Key (in case you use algorithm) to avoid theft of data.

5.) Software Development in cloud:

There are Various type of clouds like Development Cloud, Testing Cloud, Production Cloud.
The cost associated with each cloud differs. WHY AND HOW?????
Because in each type of cloud the tools and features allotted is different.

Difference between Development and Production cloud.

Under Development cloud we usually do research and development but under production cloud actual application is produced which has already been developed in development cloud. To understand compare an Application with a Car.

Advantages of development in cloud: * simple it is more cost effective, better allocation of resources between clouds, better management of security.

6.) Environment Friendly: Less use of hardware resources because of pooling of hardware resources and LOAD BALANCING FEATURE.




CA Mohit Jain
Ph: 08053881595
Email: camohitjain7@gmail.com