Tag Archives: Deduplication

Five Points to an Improved Backups

July 4, 2016 Callum Huddlestone

It is a well documented fact that backups fail at times even though they are considered critical for business continuity. Such failures happen when companies fail to keep a copy of the data outside in an offsite, remote location, away from the current network. IT admins and managers could avoid such issues with carefully pre-defined planning.

Frequent Backup Test

It is very common to find out that backups that are performed seemingly with out error messages, but later end up being unreadable when it is time to recover the data. Test the backup systems frequently in order make sure that backups are happening without any issues.

Test the files, folders, systems and servers regularly. You can run the backup tests once a week for a randomly chosen single file, once a month for a random directory and yearly for servers. Companies using multiple backup locations (for instance, cloud backup and appliance based, etc.) are required to test both systems, in the cloud and onsite.

Use Updated Technology

Newer backup technologies are always being developed; the newer versions are getting better and better and while the costs remain fair and reasonable. It is very important for the users to communicate with their service providers in order to learn about the backup servers, systems and networks. This will give the users an opportunity to avoid problems in data encryption and recovery. IT managers may utilise some applications like deduplication to reduce file sizes.

Deduplication is a well known process that helps to create duplicates of files with smaller size. This process is also effective because it enables the system to capture more snapshots in less time. It also further reduces the time period between snapshots.

Apply Systematic Incremental Backups

Older backup servers and systems may utilise full backups on a weekly basis; while daily incremental backups happening every night. This strategy is useful because it enables the system to update the data every night, without losing the files. However, it is necessary to ensure that data saved will be recovered in the original form, without any corruption, making sure that incremental backup is performed. New applications can consolidate each and every incremental backup into the full backup so that backups can be restored in only one operation.

Try Snapshots & Replication

These systems are considered more helpful for the enterprises because of the high level of redundancy, speedy backup creation, data protection, and data recovery. The users can easily develop multiple backup folders with the help of snapshots taken from the systems containing different types of files and folders. These snapshots should be preserved in prescribed systems in order to create the backups. It is also important to run the recovery tests after short intervals.

Prefer the Self-Service Portals

The need for data recovery arises mostly due to human error. Most users tend to restore data by themselves with out the need of an IT admin. This means, less admin intervention is needed. Develop a portal that allows restoring all types of files or folders from the networks and systems. The self-restoring portals have become very important for this purpose. These portals typically only allow the users to restore data (without changing or deleting them), thereby, providing do-it-yourself self-service with minimal risk.

The Role of WAN Accelerators in Cloud Backups

June 27, 2016 Damien Garvey

As some cloud backup companies fold and close, many more new ones are opening doors. The cloud backup market has become very much crowded; and many are trying their best to differentiate themselves from the crowd. Some of the new entrants are offering sophisticated hybrid cloud backup services, which makes it very difficult to compare the different options. However, many buyers consider the speed of data transfer to the cloud as one of the most important features to consider.

Data transfer speed can be greatly improved by WAN acceleration.

WAN Acceleration is a technology that is used to improve the effectiveness of data flow across a Wide Area Network (WAN). On the other hand, WAN Accelerator is a software or appliance that improves bandwidth in order to make the user’s experience more effective on WAN. Many believe that the speed of data transfer is limited to the speed the Internet connection of a company. Although this is true, technologies and software of the backup vendors play a role in increasing cloud replication speed with the help of WAN acceleration. Therefore, how do WAN accelerators work? How does the vendor improve throughput? What are the things to explore in order to select the best service provider?

Here are some of the questions that companies should ask to find the best solution:

1 – How Does the Vendor Recuperate from Disruptions of Network?

Due to packet drops and network hiccups, the transfer of information is not an easy task. Thus, it is necessary that selected vendor offers an updated technology that can replicate all data to the cloud. Matured cloud backup services, that are using traffic optimisation, deploy resilient-resumption technology allowing them to quickly recover and maintain operations even when the system experiences some kind of network disturbance or packet loss.

2 – Does the Vendor Throttle Bandwidth?

Administrators use traffic technology to specify the bandwidth that can be consumed by backup traffic at different hours of the day (such as, high bandwidth during non-business hours and low bandwidth in business hours). Best vendors will use dynamic maximum transfer unit (MTU) to optimise throughput by sizing the amount of sent on each request, depending on the network connection of the company.

3 – How Does the Vendor Deduplicate the Data?

Sending data across WAN is not an effective way to accelerate data transfer. In order to improve the transfer of data across the WAN, data scientists came up with dedulication process. Duplication helps in reducing the bandwidth of WAN by more than 90%. Despite the fact, most of the vendors make use of source-side type of deduplication to make certain that the used appliance maintains single file version. However, not all cloud vendors dedupe the information over the WAN (to ensure that replicated information does not already exist in the cloud backup repository).

The best of the breed cloud backup vendors use block-level deduplication over the WAN. Such technology checks the cloud repository to enquire whether a specific block is present in cloud backup or not. In case it is available, then it is not copied to the cloud. Conversely, if the block does not exist, it will be copied and transferred to the cloud.

You will be a well informed cloud backup consumer if you ask the above questions on network optimisation and WAN acceleration.

eDiscovery- Fulfilling Legal Mandates

June 29, 2015 Callum Huddlestone

“Data Search” and e-discovery are terms that spring up often and often in the context of compliance to legal mandates. How does cloud computing facilitate the process? Here are a few insights into what cloud service providers do to help customers keep their data discoverable for legal purposes.

Cloud computing challenges all assumptions about physical boundaries. This increases the need to protect the data and secure the privacy of individuals who entrust their personal information to business entities. This drives the need for new types of metadata that can ensure data integrity, security and privacy of the individual in an increasingly virtualized environment. The new metadata must create new types of relationships with content in an environment where sharing is integral to computing. New supports and safeguards must be evolved as new dangers emerge and evolve if legal mandates have to be fulfilled.

Electronic data in the cloud may take many forms. These include evidentiary material in the form of telephone logs, transactional data, emails, instant messaging traffic, all manner of corporate communications, documents, records, video and sound recordings, and any other kind of relevant textual or non textual information. All this material must be securely stored, easily accessible, formally indexed, uniquely identified and e-discoverable. There must be a common, flexible, extensible framework that facilitates deduplication, conversion, evaluation, summarization and storage of this information in retrievable electronic formats.

Cloud software systems store all the above information in encrypted formats and generate metadata containing document properties for ease of discovery. Metadata is data about data and is a time tested system for data discovery in the world of computing. Complex metadata is becoming the foundation for targeted advertising as well as mandated legal data discovery systems. Analytical support for this process may include limited entity extraction (eg email header) for pattern matching purposes and to cluster documents according to identified topics. Review solutions may permit user defined document sets with time based selection controls. Text analytics may discern important entity relationships and extract these features for classification and analysis of documents in storage.

We, at Backup Technology, believe that the legal discovery challenge is not in the data storage, but in the generation of metadata. Millions of users with trillions of files and billions of generations can recover the smallest bit of data from the cloud server if the metadata is properly generated and securely stored in the system. Backup Technology has proven the potential of its technology by ensuring that the right kind and quality of metadata is generated every time a bit of data is uploaded to its servers. Our team knows how it can be scaled up, reliably stored and quickly discovered.