Amazon has been growing larger, and as a result the size of their Oracle databases have been increasing in size as well, as does the number of databases they have to maintain. This has caused them to begin backing up legacy Oracle databases to tape, and led them to consider alternative strategies – one of which is utilizing the cloud services such as AWS (Amazon Web Services). Some of the challenges that they faced were: utilization and planning for capacity, tape hardware, the needed data center space, and enterprise licensing fees for tape software. They also needed highly skilled staff to spend time with setup, certification, and engineering archive planning, instead of focusing on higher value endeavors. At the end of every fiscal year, Amazon had to spend a lot of time projecting future capacity requirements – which required time consuming audits, forecasting and budgeting.

Tape robots provide some rudimentary features, but to get the most out of them Amazon had to invest in proprietary backup software. The high cost of this software added significantly to their overall costs, and it was always a problem with their budgeting but it was difficult to do away with it, because backups always needed to be written to tape devices. Maintaining reliable backups, on top of being fast and efficient when retrieving data requires a lot of time and effort with tape. When it data needs to be stored on tape, multiple copies are required. If there is a hardware failure, technicians are going required to restore the data. Contention for tape drives from multiple users requests slows down the restore process even more. (Amazon Web Services 2018)

Amazon evaluated Amazon S3 for economic and performance improvements for data backup. During that evaluation they looked at the security, availability, and performance aspects of Amazon S3 related to data backups. Amazon also conducted a cost-benefit analysis to decide if migrating to Amazon S3 would be financially viable. The analysis found that there was a performance advantage and cost competitiveness. Amazon thought it would be important that overall costs of the backups did not increase. Amazon also required faster backup and recovery performance; the time and effort required was seen to be a vast improvement over using tape. Restoring from Amazon S3 ran for two to twelve times faster, than a similar backup from tape.

Amazon also noted that there was an increase in durability and availability, Amazon S3 is designed to provide 99.999999999% durability and 99.99% availability of objects over a given year. (Amazon Web Services 2018) They also noted that AWS met all of their requirements for physical security, accreditation, processes and data protection (in-flight and at rest), as well as utilizing suitable encryption standards. Amazon DBAs determined that using Amazon S3 for backups was easy to implement because it works perfectly with Oracle RMAN.

Amazon has also found several benefits to using Amazon S3 such as elimination of complex and time-consuming tape capacity planing. Another benefit was reduced capital expenditures, which means Amazon no longer needs to acquire tape robots, drives, inventory, data center space, networking gear, enterprise backup software, or prediction of future tape consumption. This also removes a chunk of the time spent working on their budget. AWS has enabled Amazon to keep pace with the rapid expansion that comes from every acquisition they make. With Amazon S3, teams simply pay for what storage they use, and are billed as they consume more.

This also means that data is available for immediate restoring, and there’s no need to locate or retrieve tape. If a DBA needs to restore data from tape, he or she is facing a delay. The tape backup software needs to read the catalog, find the right tape, mount the tape, and read the data from it. In almost every single instance of this, the data is spread across multiple devices, resulting in delays.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s