All the speed and protection you need, without the ticket price

All the speed and protection you need, without the ticket price

Flash changes the game for storage. But how can you ensure that data protection doesn’t become a costly bottleneck for your flash workloads? Let’s find out how you can get the most from your migration to flash by making data protection an application-integrated function of flash storage.

Risk mitigation is a critical imperative in the all-flash data center

The transition to the all-flash data center typically involves the consolidation of multiple storage arrays to a single all-flash array that is handling hundreds of thousands of IOPS. In addition, SSD drive sizes are doubling year on year and technologies such as deduplication are sharing application data. 

There is therefore a higher risk profile for failing infrastructure in all-flash environments. Higher application density per TB, as well as more hosts and VM’s equate to a larger failure domain.

Ensuring application protection in all-flash environments is therefore an absolute necessity.  But should you protect data the same way when it is sitting on a flash array as you did when it lived on HDDs?  Put another way, will your existing data protection schemes withstand the data protection recovery time objectives (RTOs) and recovery point objectives (RPOs), Service Level Agreements (SLAs) and application performance demands of an all-flash primary storage environment? How can you ensure that data protection doesn’t become a costly bottleneck, impacting the efficiency of your flash array and the applications stored on it and thus negating the benefits of primary storage flash?

Flash is changing the game for data protection

The limitations of traditional backup in all-flash environments

Traditional server-centric backup processes provide reliable “off box” recovery and retention.  But they can impact application performance in all-flash environments with data typically flowing through the application and backup server.  Compromising performance for protection is not an optimal outcome if you have migrated to a flash array to accelerate your workloads!

In addition, the combination of unrelenting data growth and a complex data path means that the enhanced data protection SLAs – typical of all-flash environments - are at risk if you rely on traditional backup approaches.  User’s expectations that backup and recovery speeds should match those of flash could be severely undermined.

Backup server infrastructures also introduce escalating cost and complexity with the need to disruptively acquire and manage additional servers, software licenses, and networking equipment as application data grows.

Traditional backup was not designed for the speed and efficiency requirements of all flash environments

Snapshots and replication are not backups

In high-availability virtual flash environments, snapshots and replication are typically the first line of defence against data loss.  Flash array-based snapshots and replication provide fast, non-disruptive point-in-time copies of your data. But snapshots alone cannot deliver comprehensive backup as they have retention limitations and a dependence on the underlying flash storage system. Simply put, your snapshots will be lost if the flash storage system fails.  Snapshots are thus designed to complement as opposed to replace backup.

Array-based replication provides the redundancy that enables rapid recovery from hardware platform outage or site outage. However, replication will not provide comprehensive protection against file loss or file corruption as errors, deletions and corruptions affecting the primary copy can be replicated to a second site.

Snapshots alone cannot deliver comprehensive application protection in all flash environments

Combine snapshots, replication and backups for best-practice data protection

A more effective approach to protecting data on your flash arrays is to combine the near instant, non-intrusive performance of local and remote snapshots with the reliable recovery and cost-effective retention of backups, delivered in an application-aware, flash-integrated data protection solution. 

Welcome to flash-integrated data protection!

HPE Recovery Manager Central software integrates HPE 3PAR StoreServ All-Flash Arrays with HPE StoreOnce Systems to provide a converged snapshot and backup service optimized for flash environments.

It’s augments traditional backup approaches, combining the performance of snapshots and replication with the protection of backups.

As a result, RMC gives you fast, efficient, reliable, and simple protection for your business-critical applications running on your 3PAR All-Flash Arrays. 

Benefits of HPE RMC

Rapid and non-disruptive online recovery

RMC allows you eliminate the backup window and enable rapid online recovery with hundreds of fast, non-disruptive, application-consistent snapshots.   This reduces the backup impact on your applications, which helps support the performance goals of your 3PAR flash deployment.

Protecting workloads at the speed of flash

RMC’s Express Protect feature delivers the performance that flash application SLAs require with 23 times faster backups and 7 times faster restores than traditional backup methods. 

It achieves this by enabling the backup of snapshot data from 3PAR to a physical or virtual StoreOnce appliance, independent of backup server software.   Only changed blocks are sent to the StoreOnce System, this minimizing the impact on the application host or all-flash array.  Every backup completes at the speed of an incremental, but is stored as a synthetic full backup, which makes application recovery faster and more efficient.

Less cost and complexity

In the all-flash data center, capacity efficiency is the key to controlling storage costs.  RMC streamlines and simplifies the backup architecture required for protecting applications.  All backups are deduplicated using award-winning StoreOnce Federated Deduplication technology, thereby reducing backup storage requirements by up to 20 times.

Application managed data protection

RMC allows application integrated protection for VMware, Microsoft SQL, Oracle, SAP HANA, 3PAR File Persona and any Microsoft VSS-enabled application running as a virtual machine.  VM, Oracle, SQL and SAP HANA database admins can manage application consistent snapshots, backup, and recovery seamlessly and directly from within their preferred native interfaces.

RMC also enables crash consistent backups and snapshots to be performed on any application running on 3PAR.

An API is also provided for plug-in scripting purposes.  This enables developers to integrate the snapshot management and backup benefits of RMC with their own business applications.

Reduced risk exposure

Backups stored on HPE StoreOnce are self-contained, fully independent volumes.  These can be restored back to the original or a different HPE 3PAR array in the event of a disaster, or if physical problems occur in the production storage environment. This level of data protection cannot be achieved with snapshots alone.

The bottom line

 HPE 3PAR Flash-Integrated data protection with HPE RMC and HPE StoreOnce provides application owners with flash speed protection while reducing cost and complexity.

So when you invest in a new 3PAR StoreServ Flash array, don’t overlook data protection. A flash integrated data protection solution is critical for getting the most from your migration to flash.  If flash is the future of storage, then the future of data protection is flash-integrated backup!

Find Out More

HPE Data Protection - All the protection you need - wherever you need it

Simon Watkins

Simon Watkins  Simon Watkins

WW Product Marketing Manager, Backup, Recovery & Archive (BURA) HPE Storage

Other posts by Simon Watkins

Contact author