Axxana briefed European analysts in June about its disaster recovery solution that captures all write operations even if the primary site suffers severe damage.

The Disaster Recovery Dilemma

Most IT professionals appreciate the philosophy of the ideal DR blueprint: by putting a significant distance between two or more interlinked IT sites. Should one site suffer an incident like fire or flooding, then the other site(s) will take over. In reality though, most businesses settle for a less than perfect DR architecture due to cost. Distance furthermore imposes a synchronisation penalty with most IT remote-copy designs settling for asynchronous replication instead of synchronous replication. Or put another way: customers who don’t go for synchronous replication must accept the risk of losing buffered or uncommitted writes.

Enter Axxana

The Axxana solution and philosophy is engineered around its Phoenix BlackBox, which stores and protects the uncommitted updates of the primary site. In the event of a severe incident at this site, these uncommitted updates can be retrieved through wireless communication from the Phoenix BlackBox through secure encrypted communication. And at significantly lower cost.

The Phoenix BlackBox

The solution encompasses:

  • A fireproof container containing storage, electronics & communication
  • Axxana Collectors & Recoverers to facilitate updating the remote SAN with uncommitted updates
  • Software managing the Axxana solution

The Axxana YouTube videos adequately demonstrate the fire resistant qualities of the fireproof container and the ability to communicate wirelessly for a considerate amount of time.

At present the Axxana solution is working with EMC CLARIION & EMC RecoverPoint. But we would expect Axxana to be in the process of developing similar solutions with other vendors.

The Axxana Value Proposition

Traditional DR philosophy only recognised two separate remote copy scenarios: being either synchronous or asynchronous. But the synchronous philosophy is compromised when distances exceed 60 miles between sites because performance degrades with distance. Axxana has identified a third scenario where data is not lost despite severe damage to the primary site in an asynchronous setup.

The Axxana proposition is appealing for two reasons:

  1. To asynchronous replication customers who desire ‘no lost buffer’ IOs also known as RPO of 0 (RPO: Recovery Point Objective)
  2. To synchronous replication customers where performance degradation by DR adversely affects business applications

Accordingly for either types of customers:

  1. Change the DR design philosophy to recover outstanding replication transactions with Axxana Phoenix to be applied to the secondary site
  2. Switch to an asynchronous DR model as in Point 1

In both scenarios, the asynchronous model improved by Axxana will suffice for most sites and at a lower price than the traditional synchronous design.

Considerations

Most companies could do a better job at DR, and most IT professionals know what they would like to improve. The Axxana solution will undoubtedly be a welcome additional option for IT people who have struggled with the traditional DR design philosophies.

A selection of good local Axxana case studies should drum up interest at sites where RPO cannot be compromised. As well as for multinational businesses who are using remote copy by linking IT sites from one continent to another.

Cost is a natural barrier to implementing state-of-the-art DR solutions and many businesses will absorb downtime as the IT environment is re-established. But the general trend is towards raising the bar for DR implementations.

The Axxana design model clearly needs to be implemented with care as well as being well understood and of course being tested rigorously. But that applies to all DR implementation as well as the outsourced versions.