How to Reduce Your Data Storage Costs with Veritas NetBackup on AWS
By Dan Taoka, Partner Solutions Architect, ISV Workload Migration Program – AWS
By Henry Axelrod, Principal Solutions Architect – AWS
By Mo Hasan, Global Technical Alliance Manager – Veritas
Migrating your Veritas protected data, tape backups, and infrastructure to Amazon Web Services (AWS) allows for increased flexibility in managing data through its lifecycle.
Reducing costs while keeping your data resilient is a challenging task. The need to store data for backup, disaster recovery (DR), or compliance purposes means that multiple copies of your data are being stored within your environment.
Leveraging this data lifecycle management strategy in conjunction with Veritas NetBackup‘s deduplication capabilities allows you to realize cost savings over managing your data on-premises.
As part of the AWS ISV Workload Migration Program, Veritas has had their migration methodology validated and leverages the program for technical enablement, investment, and go-to-market benefits and support.
This post explores the options available when migrating your data and/or Veritas infrastructure to AWS and the realized cost savings.
Veritas is an AWS Storage Competency Partner that helps enterprises address information management challenges, including backup and recovery, business continuity, software-defined storage, and information governance.
Veritas on AWS Use Cases
There are a number of options for enterprise customers looking to leverage Veritas on AWS—from customers that run all of their infrastructure on-premises to those that are all-in on AWS, and everything in between.
Let’s explore the different customer footprints and use cases for leveraging Veritas on AWS.
Customers that are 100 percent on-premises are a good candidate for using any one of the Amazon Simple Storage Service (Amazon S3) storage tiers for DR and long-term retention.
Figure 1 – Veritas NetBackup with AWS.
This deployment model allows you to continue to use your data center for all day-to-day activities, but AWS becomes the home for all of the backup and archived data.
The footprint for the AWS services used is relatively light:
- Amazon Elastic Compute Cloud (Amazon EC2) for data processing and data storage catalogs.
- Amazon S3 for data storage.
- Backups are moved from on-premises to AWS in a variety of ways, including over AWS Direct Connect, over Amazon VPN, or over the internet. Depending on the amount of initial backup and archive data that needs to be migrated, the same connection is used or larger amount of data (100+ TB) are migrated with an AWS Snowball.
In addition, if you’re only looking to store your data on AWS without any accompanying infrastructure, Veritas NetBackup is integrated with AWS services at an API level and sends data directly from on-premises to S3 storage tiers without the need for any NetBackup components running in AWS.
This allows you to leverage S3 for storage without the additional overhead of additional AWS services.
Hybrid customers that split their infrastructure between on-premises data centers and AWS can leverage Veritas for backup and disaster recovery.
Figure 2 – Automated DR to AWS with NetBackup and Veritas Resiliency Platform.
This deployment allows customers to not store their backup and archive data in S3, while also leveraging NetBackup resiliency to have a standby copy of the on-premises infrastructure available on AWS in the event of a data center failure or natural disaster.
This hybrid model also allows you to use AWS Outposts, running AWS infrastructure in your own data center. Veritas has been certified to run on AWS Outposts, and has additional information on Application Availability for AWS Outpost Service.
100% on AWS
Customers that are running 100 percent on AWS are able to leverage Veritas InfoScale on AWS for backup and high availability (HA).
InfoScale provides seamless migration of critical applications like SAP, Oracle, and MSSQL from on-premises to AWS.
Figure 3 – High availability with primary storage – InfoScale for AWS.
This deployment model provides an enterprise-grade storage and backup solution by replicating these applications and their data across multiple AWS Availability Zones (AZs) ensuring high availability.
This is also a solution for customers looking to migrate all of their infrastructure to AWS from their current on-premises environment.
Managing Your Data
Whether due to security or compliance reasons, and regardless of whether the data is still deemed useful or not, your data needs to be backed up and archived.
Developing a data lifecycle strategy that defines the process and policies that govern the data as it moves through various stages, from creation to deletion, allows you to have access to data when you need it. This also ensures data is archived for compliance and regulatory purposes.
Implementing your plan on AWS to leverage the different options for storing data on S3 also helps to reduce spend.
Figure 4 – Amazon S3 storage classes.
The best way to manage this is to write the data to all of the applicable data tiers, and then expire it from the appropriate tier as time progresses.
For example, if a data file is created and needs to be immediately accessible for six months, accessible within minutes for two years, and stored long-term for compliance purposes for 10 years, you write the data to Amazon S3 standard, Amazon S3 Glacier, and Glacier Deep Archive upon creation.
The file is then set to automatically expire from S3 after six months, from S3 Glacier after two years, and from Glacier Deep Archive after 10 years or to not expire at all.
Organizations face the challenge of having multiple copies of the same data residing in their environment. This includes various copies of backups, both structured and unstructured, or copies that have been made over time.
All of these copies take up storage space; leveraging NetBackup’s deduplication helps to reduce the amount of data that’s stored, reduces the bandwidth needed for backups, shortens the time to backup, and reduces the overall infrastructure.
There are various NetBackup deduplication options that allow for deduplication at various points in the backup process.
Media Server Deduplication
NetBackup clients send their backups to a NetBackup media server, which does the heavy lifting of deduplicating the backup data using the NetBackup Deduplication Engine.
The data is managed by the Deduplication Engine and written to a Media Server Deduplication Pool on Amazon S3.
Figure 5 – NetBackup Media Server deduplication.
This option greatly reduces the amount of network traffic using the NetBackup MSDP client to deduplicate each user’s backup data before it gets sent to S3 for storage.
Figure 6 – NetBackup client-side deduplication.
NetBackup Appliance Deduplication
There are other options for appliances to perform the deduplication of the data. Learn more on the Veritas Appliance Support page.
Veritas’ ability to deduplicate the data reduces costs, and customers have realized savings of more than 90 percent. One customer reduced their data footprint from 2 PB to 100 TB.
The total Cost of ownership (TCO) of data protection varies by customer. By leveraging NetBackup’s deduplication to reduce the amount of data stored, using S3’s multiple storage classes to develop a data lifecycle management, and removing the manual steps of existing backup, archival, and recovery processes, you can reduce your TCO while still adhering to data protection and auditory requirements.
Using Veritas on AWS, automating the backup and archival process is easier to do than when using tape archives. This is done by creating policies to automatically expire the data from the different S3 storage classes, as it’s no longer needed.
Data, even when stored in Glacier Deep Archive, is available in hours, something that’s very difficult to do when retrieving tape backups that are stored off-site. The data is durable and reliable because it is stored in multiple Availability Zones, unless explicitly set to be stored in only one AZ, to ensure it’s available when needed.
This is a feature of S3 that require multiple copies of the data to be stored in multiple data centers if you are doing the same thing in an on-premises environment.
Using S3 for data storage also means the manual process of moving tape backups to offsite storage, and retrieval of data, is no longer needed. Combined with the deduplication of data being stored in S3, this is something that’s not typically done with tape backups and means the amount of data being stored is reduced.
A high-level example of TCO savings is seen if we look at the cost of storing tape backups versus migrating your archive data to AWS.
- Let’s say the cost of your 30TB LTO-8 tape for backup is $100 per tape; this breaks down to $3.33 per TB stored.
- The same amount of data is roughly $4/TB per month when stored in Amazon S3 Glacier, but the cost goes down to $0.99/TB per month when the data is stored in Glacier Deep Archive.
- One of the biggest factors when moving your Veritas data to AWS is the act of deduplicating the data to reduce the footprint. If you have 100TB of data in your data center that needs to be stored, backing up to tape costs $333.
- If we take a conservative deduplication percentage of 90 percent, and we store the same data in S3 Glacier Deep Archive, the cost per month of storing the same data (10TB of deduplicated data) is only $9.90.
- There will be additional ingress and egress costs of moving the data to AWS, but those costs are nominal in comparison to the additional costs of data center space, worker time and productivity, the off-site archive service itself, and tape library maintenance.
- This also does not take into account the fact that tapes degrade over time and the ones being stored offsite need to be moved back onsite for restoration.
Peerless Clothing is a producer of fine men’s tailored clothing that has benefited from migrating to AWS.
By migrating their Veritas backup and disaster recovery to AWS, Peerless Clothing was able to reduce the amount of data that needed to be archived from 315 TB to 16 TB—a 95 percent reduction.
Data came from a variety for sources, Oracle, SAP Hana, Microsoft SQL Server, VMware, Windows, and Linux, which was processed by NetBackup, and moved to S3 for storage.
Figure 7 – Peerless Clothing reference architecture.
Peerless Clothing also used Veritas Resiliency Platform for disaster recovery and migrating portions for their infrastructure to AWS. This allows for flexibility in their development and testing of internal applications.
Peerless was very pleased with the joint solution as they used NetBackup to migrate their workloads and AWS for both DR and long-term retention. This solution was cost-effective and allowed Peerless to leverage the cloud to simplify their backups and DR and in a way that was not possible on-premises.
“Deploying Veritas with AWS is easy, and knowing that we can quickly recover in the event of a disaster is a win for us,” says Laura Ghinda, Linux Administrator at Peerless Clothing.
Having your data available when you need it, whether for day-to-day activities, recovering from a disaster, restoring a local data loss event, or for compliance purposes, is extremely important for companies of all sizes.
Veritas on AWS ensures your data is highly available and provides savings over storing data in your data center and/or managing the tape backup archival process.
Veritas – AWS Partner Spotlight
Veritas is an AWS Storage Competency Partner that helps enterprises address information management challenges including backup and recovery, business continuity, software-defined storage, and information governance.
*Already worked with Veritas? Rate this Partner
*To review an AWS Partner, you must be a customer that has worked with them directly on a project.