Japan, a Painful Reminder about Disaster Recovery

As we begin to move on from the shock of the destruction wrought by the Japanese earthquake and tsunami three weeks ago, we begin to focus on the recovery, both in human terms and in terms of business.

The latter note – disaster recovery and business continuity – has been addressed by a number of top technology publications in recent weeks. Chris Preimesberger at eWEEK called the Japanese earthquake and tsunami a “Wakeup Call for IT Managers” while his former eWEEK colleague and long-time tech media veteran Wayne Rash discussed “Truly Preparing for the Worst” over on CTO Edge. Both articles send a clear message – disaster recovery plans not only need to be installed, but also tested and updated.

Prior Planning

In his article, Rash questions whether corporations in the U.S. are anywhere near as prepared as those in Japan and notes that those businesses that have failed to prepare for such a disaster are “toast.” He concedes however, that there are certain steps, which if taken, may enable a company to emerge from a disaster and allow them to conduct some semblance of business until normalcy returns.

Some of the criteria he mentions for preparedness include having “an operations site somewhere away from wherever your company is located” and having the data stored remotely or “in the cloud” as long as “the cloud computing service provider is a long way away from where you are.”

While the storing of data remotely has become a staple of disaster recovery plans, if the data is located “a long way away from where you are” then retrieving it means going over a WAN connection, which poses problems with data recovery…or does it?

The WAN Obstacle

Even with deduplication, recovery processes involve the transfer of large amounts of data across a WAN, requiring a significant amount of WAN bandwidth. Moreover, the transfer of these large amounts of data must be done without interruption. Yet WAN latency and packet loss can disrupt this process, resulting in missed recovery point objectives (RPO).

However, WAN performance does not have to be an obstacle to disaster recovery. Use of advanced WAN deduplication – the kind pioneered by Silver Peak – enables more data to be protected across the WAN without adding costly bandwidth. Latency and packet loss also need not adversely impact backup and recovery processes when a company utilizes features such as those in Silver Peak’s Network Integrity and Network Acceleration applications.

Silver Peak’s WANop Advantages

WAN optimization (WANop) is an indispensible component of offsite disaster recovery projects and Silver Peak’s WANop solutions increase the performance and reliability of backup, replication and recovery while reducing IT costs. At a time when a company can ill-afford an issue with retrieving its data and compounding an already difficult situation, Silver Peak’s technology provides a smooth road to disaster recovery that includes:

  • Improved data transfer times: Silver Peak’s Network Memory recognizes repetitive information and delivers it locally.  In addition, Silver Peak provides enhancements to accelerate TCP and CIFS, reducing the impact of latency on data transfers.
  • Maximized WAN efficiency: Network Memory can reduce as much as 95% of WAN traffic by eliminating the transfer of duplicate information.  Network memory fingerprints at the byte level, enabling Silver Peak to detect and eliminate repetitive patterns even when the backup/replication solution is performing similar functions at the block level. In addition, Silver Peak leverages advanced compression techniques to further reduce the amount of WAN bandwidth required for backup and replication.
  • Reduced packet loss and delivery errors: Silver Peak reduces the impact of both packet loss and jitter that occurs when router links are oversubscribed and drop or re-order packets.  Adaptive Forward Error Correction (FEC) for example, can reduce effective packet loss by an order of magnitude.
  • Guaranteed data security: Silver Peak NX appliances use hardware-based AES encryption to protect network traffic and local content, protecting all data at all times from unauthorized access.
  • Cost effective scalability: Silver Peak can support a full 500 Mbps WAN capacity in a single NX appliance, with several variants available that support over 45 Mbps.  This enables enterprises to support data center to data center links cost-effectively, while also supporting network-based backups across many remote locations.

The Silver Peak solution cost-effectively scales to support large data center environments, delivering the most LAN and WAN side throughput for disaster recovery environments.  Only Silver Peak optimizes both TCP and non-TCP traffic, improving data transfer times and maximizing WAN efficiency across all backup and replication solutions.  And when disaster strikes your business, the faster you can get access to data, the faster you can get back to something resembling “business as usual” for your customers.

About the author
Jonathan Bloom