executive summary of the three research papers: (1) Holism vs. Reductionism, (2) SDLC Models and (3) Data Backup/Disaster Planning.

Table of Contents Introduction 3 Database Backup 3 Disaster Recovery 5 Disaster Recovery and Database Backup 8 References 10

Database Backup Vs. Disaster Recovery

This paper analyses the concepts of database backup and disaster recovery. Database backup and disaster recovery systems are two important arenas of business resilience and business continuity management. Traditionally, a database backup plan used to be sufficient for a disaster recovery plan. Due to the advancement of technology and complexity of businesses, database backup is now not enough for a disaster recovery plan (Watters & Watters, 2014).

The paper primarily focuses on the concepts of database backup and disaster recovery. Gradually, the concepts will be used for a comparison of the two.

Data is a very important resource for running a business successfully. Loss of business data can severely impact a business and may end up loss of revenue. Databases are used for storage of business data (Watters & Watters, 2014). A database backup is a copy of a data and files for reconstructing a database if there is any failure or unavailability of the database. There are various online and offline techniques for keeping database backups (Guise, 2008). Based on the business requirements, importance of data and various other business specific parameters, a suitable database backup and data recovery planning is executed. There are two types of database backups, physical and logical backup.

When physical data files are stored for recovery of databases then those are called physical backups. Examples of such physical backups are backups of control files, data files, ‘redo’ logs and so on. These are mostly metadata that gives information about the physical storage of databases. Usually, physical backups are taken on offline disks and tapes and are kept separated from the main database (Guise, 2008).

When logical data like the stored procedures and tables are exported from a database and stored into some binary file by using some database management software then those are called logical backups. For example, Oracle supports logical backups of data (Guise, 2008). logical backups can be re-imported into the original database after a successful export.

There are various techniques for taking database backups. Each of these has own set of advantages and disadvantages based on the trade-offs of cost and performance. The ‘cost’ of a database backup technique considers the cost for the time and storage (Guise, 2008). Different types of database backup techniques are,

1. Normal or full backup

2. Incremental backup

3. Differential backup

4. Mirror backup

Normal or full backup is the initial backup strategy for all other backup techniques. All data of a database are backed up in a full backup plan. To ensure highly-reliable database backup solutions, it needs to plan full backups frequently (Guise, 2008). Restore operation is full backup is faster than the same in other types of backup plans. For small to moderate volume of data, full backup is a feasible solution.

In an incremental backup strategy all files that have been updates since the last backup, are stored back. It takes the least amount of time. During a restoration, it needs to process each incremental backup. Hence, the time taken for restoration process will be longer. Incremental backup planning is suitable when there is a large number of files that needs to be backed up each time. It also consumes lesser storage space and supports frequent backups (Snedaker, 2013)

In a differential backup, all updated files from the last full or initial backup are taken into consideration. The restoration time is shorter than the same for incremental backup. The size of the storage for differential backup will grow significantly if there are frequent backups. The restoration process is also simpler than the same in an incremental plan.

In a mirror backup plan, all files that have been updated since the last full backup are included. However, the missing files are removed from the set of backed up files. It offers customization for inclusion of files for a backup (Snedaker, 2013). Only the selected set of files and folders will be copied into a backup data set. This is the fastest backup process compared to the other three processes. It does not support compression during taking backup of databases.

A disaster recovery plan or system of a business is a part of the business continuity plan. In early days, disaster recovery systems were synonymous to the database backup systems. However, in modern business environment, the scope of disaster recovery has been expanded (Snedaker, 2013). The main objectives of a disaster recovery plan are business resilience and business continuity. A disaster recovery plan must ensure that there will be negligible or zero data loss and recovery time in the event of a disaster or disruption. A disruption or disaster can be natural disasters like flood, earthquake, fire or technical disaster like crashing of disks, unavailability of networks, loss of data from disks and so on (Wai-chi Fang, Muhammad Khurram Khan, Kirk P. Arnett, Heau-jo Kang, & Ślȩzak, 2010).

Traditional DR systems used to be dependent on legacy systems and were less frequently tested. As a result, the backup windows were longer, tape-incentive, used to take long time for recovery, used to lack recovery objectives and most of the DR processes were manual. Such drawbacks had negative impacts on the productions.

There are two important parameters for assessment of a disaster recovery plan or system. The parameters are, RTO or Recovery Time Objective & RPO or Recovery Point Objective.

The RTO refers to the duration of the time for the restoration of a business process after a disruption or disaster in order to avoid the unacceptable consequences from the break in business resilience and business continuity (Wai-chi Fang et al., 2010).

The RPO refers to the time for the interval during a disruption. During this time an allowable amount of data may get lost. The allowable amount of data is the threshold mentioned in the business continuity plan (Wai-chi Fang et al., 2010).

A business continuity plan may be planned with aggressive RTO and RPO for meeting the targets. In such cases, the business operations are tending to be expensive enough. Some characteristics of a disaster recovery plan are,

· A set of clearly defined RTOs and RPOs.

· Consistency between the disaster recovery plan and corresponding database backup plan

· Frequent testing of the plan for ensuring recoverability of data in an event of disaster or disruption (Wai-chi Fang et al., 2010).

· Keeping logs of the changes

· Synchronization of the disaster recovery log with the other information systems for initiating a fallback effectively.

· A disaster recovery plan must be easily scalable and reliable

· A disaster recovery plan should have redundancy in it.

· It must conform to the standards and other laws and regulations.

Figure 1: DP Planning [Taken From:www.greenlight-itc.com]

Disaster recovery is a continuous process. It must be tested frequently and must be updated to meet the changes in the business environment and business requirements(Wai-chi Fang et al., 2010). The outline for a disaster recovery plan should include the following steps,

1. The goals of a disaster recovery plan must be designed for specific use cases that are targeted by the plan.

2. The DR plan must provide solution for end-to-end recovery of data.

3. The tasks of a DR plan must be specific and concrete.

4. A DR plan must implement a set of control measures for minimizing the chances of occurrence of a disaster or disruption.

5. A DR plan must be integrated with the information security mechanisms of the business.

6. The licenses of the systems and software used in the DR plan must be kept updated timely.

7. A DR recovery plan may have more than one recovery path for data

8. A DR plan must be tested regularly.

Disaster recovery and database backups are not similar. However, in many instances, these two concepts are used synonymously. Many businesses focus on only their database backup strategy and consider this as the sole disaster recovery plan of the business. There are some subtle differences between the two concepts.

A disaster recovery plan must ensure that there will be zero data loss or downtime. A database backup plan may fail during the execution of a DR plan. If there is no coverage for such issue in the DR plan, then the sole objective of the DR plan will be violated (Watters & Watters, 2014). Hence, DR plan is not only a database backup plan. It is more than that. A DR plan covers the required actions to be taken in the face of failure of a database backup plan also.

A database backup may spread backed up data over multiple sites. The main objective of a database backup is to keep the copies of a database so that the database can be reconstructed at the time of a disaster (Watters & Watters, 2014). Now, restoration and reconstruction of the database may be time consuming enough to create severe downtime for a business. On the other hand, a suitable disaster recovery plan must be there to ensure such a strategy that will help to minimize the downtime (Watters & Watters, 2014).

Database backup processes are the initial steps of the execution of a DR plan. It keeps the means to reconstruct a damaged or lost database during the execution of corresponding DR plan (Watters & Watters, 2014).

Guise, P. d. (2008). Enterprise Systems Backup and Recovery. New York: CRC Press.

Snedaker, S. (2013). Business Continuity and Disaster Recovery Planning for IT Professionals. New York: Newnes.

Wai-chi Fang, Muhammad Khurram Khan, Kirk P. Arnett, Heau-jo Kang, & Ślȩzak, D. (2010). Security Technology, Disaster Recovery and Business Continuity. New York: Springer.

Watters, J., & Watters, J. (2014). Disaster Recovery, Crisis Response, and Business Continuity. New York: Apress.