A dependable way to keep data safe in real-time

A dependable way to keep data safe in real-time

The concept of business continuity involves planning and preparation in order to ensure that an organization can continue to operate in the event of a serious incident or disaster, and can return to an operational state within a reasonably short time. After all, every business concern operates on the principle that the business will continue and grow to newer heights in the future. However, a central element in this continuum and growth is the data that any organization develops in a variety of sizes, shapes and forms. Protecting this data and ensuring that it is available to the relevant users at all times, represents an integral part of almost any IT budget within an organization. And this is where a proper disaster recovery solution comes into play. While the modes and mediums of disaster recovery are many, we are reflecting upon a more centralized approach here, where in you can protect and retain access to your data in real time.

Why you should consider array-based replication

Once the cost of downtime has been ascertained, having a proper disaster recovery solution – whether data replication or data protection or both – becomes relatively easier. While there are different modes of data replication and different elements which govern the applicability of these modes, array-based replication lies at the core of hardware-based replication. Operational at the array level, this replication solution offers few distinct advantages:

  • Workload balance – As the array handles the tasks of replicating to a secondary location, the workflow from hosts via network to storage is relatively streamlined and free from contention.
  • Application aware – Methods could be many – snapshots, volume or consistency groups however, the idea is always the same – to create an image of a data set / incremental data set without disrupting user access to the application(s).
  • Centralized structure – Depending upon criticality, data pertaining to one application, multiple applications or an entire array can be replicated to a secondary array, provided other conditions for set up are adhered to.
  • Scalability – Most array-based replication solutions can be scaled even at a later date.
  • Failover – Since both or more arrays are in sync in real-time, failover in the event of a disaster is more often instant with zero or minimal data loss, without any manual intervention.

One of biggest challenges of this solution is the need for identical solutions from the vendor on both sites, entailing configuration sets, OS version, products from the same family, etc. At times this could be a huge challenge, as it may result in a significant addition to the bill of material for the set-up.

How array-based replication works

At a broader level array-based replication works in two ways – synchronous and asynchronous.

How array based replication works syncThe synchronous mode works more in real-time; in the sense that each data packet is dealt with in tandem on both the primary and secondary sites, after which the next batch of data packet is dealt with. This mode is advisable when dealing with mission-critical applications where downtime cannot be tolerated. Given the low latency requirements, hardware is needed which is then supported by requisite licensing and support costs, hence the overall cost of deployment for synchronous replication can be higher. However, as this solution is tailor-made around keeping performance intact, the hardware / software costs involved in supporting that performance becomes immaterial, at times.

How array based replication works asyncThe asynchronous mode works in a way where a data packet is first written on the primary site and then on the secondary site. Application performance can be improved with async mode, as the next batch of data packets does not have to wait in line until the first batch is replicated to secondary site and an acknowledgement is received. The cost of deployment of async mode is relatively lower when compared with sync mode and smaller installations can benefit from this mode. However, in the event of a failure, the time it takes to get the site up and running could be relatively longer, as incremental data has to be rebuilt from the log files, which makes this mode less recommendable for mission critical applications.

How Fujitsu ETERNUS Primary Storage can make a difference

One shoe does not fit all. Likewise one replication solution which is custom made for a large enterprise may be less than ideal for a small and medium business. There could be budgetary constraints, challenges in managing the infrastructure, relatively steep upgrade costs at a later stage… take your pick. On the other hand, the basic concept or principle behind working mechanisms of disk-based replication cannot be altered either. So then the bottom line is how to make the best of what we have, in order to make sure that the end product can accommodate many use cases. This is where Fujitsu ETERNUS Primary Storage Arrays can make the difference in your data center. Let’s see how:

  • Packaged differently – Apart from having a replication solution in place, if you are also interested in performance and consolidation from your storage array, ETERNUS has an offering for you, with a deployment of ETERNUS AF on the primary site and ETERNUS DX on the secondary site. You can have a mix of replication solution backed by an array that can cater to your IOPs requirement on the primary site and another array that can give you the benefits of performance with high performing drives (SSD’s) and savings with cost efficient drives (SAS, NL-SAS) on the secondary site.
  • Synchronicity, not a mandate – Whatever your solution, ETERNUS AF/ETERNUS AF or ETERNUS DX/ETERNUS DX or ETERNUS AF/DX on the primary and secondary sites, you can have different configuration sets, system generations, OS versions, and model types (ETERNUS AF 250, 650 with ETERNUS DX 200, 500, 600, etc) on both sites. This gives you the flexibility to start small and grow at a later date, not to mention lowering your cost of deployment.
  • Manageability – Since the line of communication remains common between whatever model from the ETERNUS family that you deploy (on the primary and secondary sites), with ETERNUS SF; the result is an enhanced ease-of-use. This can be witnessed in array level management, data migration, moving volumes across, LUN enhancement and other routine nitty gritty workings.
  • Manual control – In case of planned downtime, an administrator-initiated failover option supports business continuity.
  • Automated control – For unplanned downtime, protocols ensure protection against hardware, network or environmental failure.
  • Transparent failover – Ensures business continuity, making sure data is available for application(s) / user(s) as and when they need it. This also helps to ensure zero downtime and eliminates loss of any kind to your business.

For a detailed explanation on how array-based replication works on ETERNUS Primary Storage Arrays, please refer to the Storage Cluster white paper.

Up to this point, we’ve listed some of the advantages Fujitsu ETERNUS Primary Storage brings to the table. But now lets see what our customers have to say – because they use our solutions on a daily basis and have a good impression of what ETERNUS is able to do for a business.


“The transparent failover solution was the main factor in our decision to work with Fujitsu. It would enable our services to continue running uninterrupted in the event of an error. That was important to us.”
Markus Handschuh, Head of IT System Operations, REWE DORTMUND

A brief snapshot of customer requirements and solution:

REWE DORTMUND: REWE DORTMUND Großhandel eG is an independent, regional cooperative under the REWE Group umbrella. Its almost 300 supermarket outlets supply some 512 markets in the Germany. The company focuses on locally made products in order to support the local economy, secure jobs and keep transportation distances as short as possible.

Challenge faced: The existing storage cluster solution was becoming difficult to operate given the capacity expansion needed and the IO rates delivered vs the demands of users.
Solution provided: FUJITSU Storage ETERNUS DX500 storage cluster with transparent failover for all applications, including business-critical warehouse management and inventory control systems.

Benefits observed:

  • Transparent failover solution ensures maximum availability
  • High performance solution for business-critical applications
  • Internal cache allows applications to run on nearline SAS drives
  • 16GB Fiber Channel connection ensures high performance levels
  • At least 120,000 IOPs

Please read the detailed case study here.

dbh Logistics IT AG

“What really convinced us to go with the Fujitsu system was how cost-effective it was and that it gave us the option to employ a transparent failover solution through the storage cluster without adding additional modules.”
Rudolf Klassen, Head of IT Services, dbh Logistics IT AG

A brief snapshot of customer requirements and solution:

dbh Logistics: dbh Logistics IT AG (dbh) is a leading software and consulting service provider in the fields of customs and foreign trade, compliance, transport management, port management, SAP and data center solutions. It is based in Bremen, and its portfolio covers everything from consulting, design, development and implementation to hosting and support.

Challenge faced: dbh Logistics existing system was unable to cope with the rising demand for storage space and would have struggled in the event of a disaster recovery scenario.
Solution provided: FUJITSU Storage ETERNUS DX200 with transparent failover for disaster recovery. Result is higher value of service level agreement offerings to customers.

Benefits observed:

  • High reliability
  • Faster backups.

Please read the detailed case study here.

Walser Privatbank AG

“High availability, good performance and ease of administration were all factors that contributed to our decision to choose Fujitsu’s solution. From a commercial standpoint, the company also made us the best offer.”
Sebastian Fauland, IT Infrastructure Team Leader, Walser Privatbank AG

A brief snapshot of customer requirements and solution:

Walser Privatbank:  Walser Privatbank AG is a high-profile private banking institution in Europe’s German-speaking territories. Walser Privatbank specializes in securing and increasing the assets of its upmarket private customers.
Challenge faced: Walser Privatbank was looking for a new storage solution. While high availability and high performance were major criteria, the solution also had to be cost-effective.
Solution provided: FUJITSU Storage ETERNUS DX200 with Storage Cluster for transparent failover.

Benefits observed:

  • High availability
  • Transparent failover with no service disruptions.
  • Dynamic distribution across storage media.
  • Time savings through simpler administration.

Please read the detailed case study here.

Zero downtime, zero data-loss and continuous availability of data is the need of the hour for any business concern. However, what is also required is a promising disaster recovery solution which is custom built as per infrastructural and business needs of the respective organization. With our Fujitsu ETERNUS Primary Storage portfolio, we have offerings for such requirements and more. Visit our website for more details on Fujitsu ETERNUS AF All-flash arrays and Fujitsu ETERNUS DX Hybrid arrays. If you’re still looking for the right Disaster Recovery Solution in general, you might find another article helpful we published recently: How The Right Disaster Recovery Solution Can Make All The Difference In Your Data Center.

Tags: , , , , , , , , , , , , ,

No Comments

Leave a reply

Post your comment
Enter your name
Your e-mail address

Before you submit your comment you must solve the following arithmetic function! * Time limit is exhausted. Please reload CAPTCHA.

Story Page