Printer Friendly

Is SMB disaster recovery really within reach?

The need for business continuity and the data protection required to support disaster recovery initiatives cuts across companies of all size, in every industry, and across states and countries. The network complexity associated with typical business continuity solutions has greatly eased over the past 5 years, and solutions are now available that have simplified earlier approaches to deliver the protection needed by larger organizations. But what of the disaster recovery needs of small to mid-size businesses (SMBs)--are they really within reach? The data protection requirements of these businesses can be just as critical as the needs of enterprise organizations, yet the disaster recovery and business continuity solutions that are contemplated by SMBs can easily be constrained due to complexity, manpower requirements and cost issues.

Can SMBs effectively create and deploy a centralized and continuously available storage infrastructure that ensures business continuity at a cost efficient price? The answer is a resounding "yes"--if approached in a phased and structured manner that ensures flexibility while maximizing the company's existing infrastructure investment.

To begin the process, it is important to create an overall data protection plan that consciously examines the current and overall corporate network infrastructure under a blinding white spotlight. Are there any single points of failure in the network that need to be addressed? Is there older technology in the network that would impede bringing information back online in case of data corruption or damage? And can access to data be provided (in minutes, rather than hours) to employees, regardless of location?

These are the questions that Thompson Hine LLP used to begin the process of creating a robust and highly cost-effective solution to address their data protection and disaster recovery needs. Thompson Hine, a law firm with approximately 350 lawyers and 600 support staff, sought a solution that would allow them to protect and manage data across its lifecycle, creating a centralized storage infrastructure to ensure business continuity.

Established in 1911, Thompson Hine today is among the largest business law firms in the U.S. with offices in Brussels, Cincinnati, Cleveland, Columbus, Dayton, New York, and Washington, D.C. For the last three years, the firm has been named as one of the Best Corporate Law Firms in America (in an annual survey of 32,000 corporate directors conducted by Corporate Board Member magazine). Thompson Hine serves some of the premier businesses in the world, including: Ford, Toyota, Goodrich, Goodyear, Eaton, Sherwin-Williams, JoAnn Stores, Verizon and KeyCorp.

Like many other companies, they rely on e-mail for day-to-day business operations, as their clients choose to communicate as much through e-mail as by any other means. The firm also has an extensive document library that contains client-centric attorney work product, and the need to protect this client data (effectively and efficiently) was the cornerstone of the project.

Understanding that access to and protection of client-critical data (both e-mail and library materials) was paramount to the firm's leadership and success, Thompson Hine sought a centralized means of ensuring that its business-critical data was continuously available. Spurred by the sudden failure of a Brocade switch in their Cleveland SAN, Thompson Hine executives decided to merge the disaster recovery plans that were in the process of development with a data protection strategy that would protect primary data locally at each office (either on SAN or direct attach storage) and would efficiently replicate data from their six offices to multiple off-site locations.

Shining a Bright Light on the Process

Examining their current network environment and assessing requirements for primary and secondary data locations, the firm's IT leadership established several main objectives as the overarching requirements needed to create a solution that would ensure business continuity in the event of data damage or disaster. These objectives included:

* Utilize the existing infrastructure as the foundation for the solution to reduce overall costs

* Eliminate single points of failure in the network

* Identify and protect critical data, deploying a solution that allowed immediate access to data as needed

* Make critical e-mail and document data available from anywhere to allow continuation of ongoing work

* Provide workspace and standby office areas in case of damage to existing facilities.

Recognizing the challenges associated with deploying a comprehensive data protection solution for multiple locations, Thompson Hine chose to roll out their solution in a phased approach. While this approach may take slightly longer to complete, it brought several advantages to the table that SMBs can appreciate. A phased deployment allows implementation in segments to minimize the cash needed up front, brings a way to take advantage of new technologies as they become available, and gives companies the ability to begin project rollouts while formalizing their plans. Most importantly, a phased approach can allow a smaller company to accomplish what needs to be achieved, in this case, freeing up IT resources to work on multiple projects while the data protection/disaster recovery deployment was underway.

Determining the Path

Reviewing their facilities for single points of failure, Thompson Hine determined that while the connectivity infrastructure (that started this process rolling) could be easily addressed with the addition of several new switches for redundancy, the Hitachi-based SAN that supported the Cleveland office opened up a level of exposure for the company if the data was unavailable. Due to the heavy concentration of employees at the corporate headquarters in Cleveland (and the vulnerability of the SAN at that location), it was decided that the first phase would address protection of this primary data by a combination of local mirroring and replication through a dedicated DR replication T1 line to a secondary data center in their Cincinnati office.

With a set of objectives, commitment to a phased approach and an agreement as to their first priority, Thompson Hine began to discuss their needs with integrators. Several weeks into the process, they met with the Chi Corporation (a leading specialist in storage networking solutions) and they presented a highly efficient and cost-effective approach that addressed Thompson Hine's needs. Chi is seen as an organization that was critical to the success of the project, with the integrator serving as the local resource that delivered the solution and providing the experience that allowed Thompson Hine to meet a very tight schedule and accomplish their goals. The selection was made, and the project was underway.

4 Days to Protect and Serve

While Thompson Hine executives negotiated with local conference centers to provide space for employees in case of catastrophic damage to facilities, IT personnel began the process of implementing their business continuity solution. To protect data on the Cleveland SAN, Thompson Hine selected a Nexsan ATABeast with 6.1TB of usable capacity to serve as a repository for their SAN data, supplementing the existing tape backup storage that was on site. The ATABeast is the first true online disk vault that delivers industry-leading features with unparalleled density and security at a Total Cost of Ownership (TCO) that is less than tape. Supplementing tape with the ATABeast not only delivered SAN redundancy with TCO advantages but also allowed the company to avoid the long restore times associated with tape libraries and bring data back online immediately for use by attorneys and support personnel.

Preparing the Cleveland data center, Thompson Hine started the first day of deployment by racking and physically connecting the hardware, configuring the system, virtualizing the ATABeast and testing connectivity and failover. Day 2 was spent preparing the Cincinnati data center, racking and connecting hardware, and configuring and testing of connectivity. Day 3 saw the installation of FalconStor's IPStor 4.0 to provide a flexible and hardware independent way to enable snapshots, and then 700GB of seed data was mirrored from the Hitachi and replicated locally to the ATABeast in Cleveland and to an ATABoy2 (which in this case was used as a storage platform that could be easily transported to help seed the secondary SAN).

Starting Day 4 of the project, the focus moved back to Cincinnati. Local mirroring was complete in the primary location and the ATABoy2 storage was disconnected and driven to Cincinnati, where the data was transferred to a duplicate Hitachi SAN in the secondary data center. Replication locally and via WAN was then started, bringing the backup to an up-to-date status. Administrators then scheduled e-mail backups to occur nightly and document library backups to occur every 15 minutes during the day to the local ATABeast, and every 24 hours from Cleveland to the secondary location, protecting the two prime data sets needed by the company. While 70GB of e-mail data and 2-3GB of document data changes every day, Thompson Hine has used the iPStor servers (not the host servers to manage replication, taking the load off of production servers and minimizing traffic on their T1 line.

Securing High-Risk Locations

Moving on to phase two of the project, Thompson Hine looked to replicate e-mail and document data from their Washington D.C. and New York City locations to the Cleveland data center. Executives saw both locations as areas that carried a higher risk for data and employees at each location were acutely aware of the value of data and potential risk faced.

As with the Cleveland installation, administrators needed to use the existing infrastructure, had only limited downtime available for implementation, and process could not impact production.

Choosing to install a local mirror for the direct attached storage in each location, Thompson Hine this time used individual ATABoy2s (instead of ATABeasts) as a data repository for each office. Both Washington D.C. and New York City were individually connected via WAN to the primary Cleveland SAN; 200GB+ of data from each location was seeded to the Cleveland SAN with the use of a traveling ATABoy2 to minimize setup strain on the WAN. Storage personnel then synchronized data on all platforms--the ATABoy2s on the East Coast and Hitachi SAN in Cleveland--and started scheduled e-mail and document backup for each location.

Achieving Data Protection With ATA-Based Approaches

With the new, cost-efficient infrastructure, Thompson Hine can now effectively replicate data overnight from multiple offices to its secondary data center through a T1 connection. Utilizing its existing production WAN infrastructure for the nightly replication of data from its satellite offices, the firm has eliminated increases in communication costs since a dedicated replication connection is not needed. The solution's ability to mirror between dissimilar storage subsystems enabled Thompson Hine to mirror its main data center and East Coast offices to a cost-correct and higher capacity Nexsan ATABeast and ATABoy2s, respectively, providing seamless failover and redundancy for multiple locations.

So what's on the IT schedule for Thompson Hine in 2005? According to IT personnel, ATA drives have proven themselves as a way to protect valuable case files, briefs and other information critical to day-to-day operation, and they're sold on the cost-correct storage technology. In addition to converting older direct attached storage in New York and Washington to newer ATA-based ATABoy2 disk solutions and upping capacities on all of the storage infrastructures in place, the firm plans to approach phase three by rolling out identical ATABoy2-based storage deployments in their Dayton and Columbus offices. They also plan to install an ATABeast to protect the Hitachi SAN in their Cincinnati data center and use the existing ATABeast in Cleveland to serve live data to users when they replace their primary Hitachi SAN later this year by simply breaking the mirror and pointing users to the ATABeast data.

With a network of offices around the world, it was imperative that Thompson Hine find a way to consolidate and replicate important files and communication, effectively managing data along its lifecycle. By using ATA-based solutions, companies of all sizes can leverage the technology's lower costs, higher capacity and superior reliability into its current high-end storage environment in order to ensure the best ROI while also maximizing data availability.

By using ATA-based solutions, existing technology and phased deployments, SMBs can implement disaster recovery and business continuance solutions without constraints that ensures the utmost flexibility investment protection. With ATA, SMBs have a way to effectively deploy a centralized, flexible and continuously available storage infrastructure that ensures business continuity at a cost efficient price.

And that's a win for every-one involved.

Brendan Kinkade is vice president of marketing at Nexsan (Woodland Hills, CA)

www.nexsan.com
COPYRIGHT 2005 West World Productions, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Disaster Recovery & Backup/Restore; small and midsize businesses
Author:Kinkade, Brendan
Publication:Computer Technology Review
Geographic Code:1USA
Date:Feb 1, 2005
Words:2036
Previous Article:SAS delivers maximum SATA scalability.
Next Article:The SMB users guide to data protection.
Topics:


Related Articles
Data grid disaster puts SMBs at most risk.
Plan for the worst, hope for the best: backup and disaster recovery.
Implementing an IP SAN for disaster recovery: using iSCSI as an enabler.
Overcoming recovery barriers: rapid and reliable system and data recovery.
Fighting SMB and nonprofit storage fires without getting burned.
Protecting Microsoft Exchange Server in SMBs.
Peace of mind: disaster recovery plans can keep your business alive.
Business continuity in SMB: disaster recovery in the small-to-medium business space.
Personal disaster recovery software: an essential part of business disaster recovery plans.
Infosecurity Europe 2007.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters