eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
With Microsoft Corp. Exchange mailboxes growing to criminally high levels, IT staffers at law firm Perkins Coie LLP had to overhaul their storage infrastructure to keep lawyers productive. Perkins Coie is the largest law firm in the Northwest and counts heavy hitters such as Microsoft, Adobe Systems Inc. and The Boeing Co. among its clientele.
Jim Ryan, systems infrastructure manager at Seattle-based Perkins Coie, wanted not only to deal with the firms data growth in its mail stores but also to minimize downtime. Ryan and his staff decided to implement a SAN (storage area network) solution built on EMC Corp.s Clariion CX arrays and SAFE (Storage Administrator for Exchange) to boost scalability and availability.
With 700 attorneys and close to 2,000 mailboxes to manage, Ryans e-mail infrastructure is fairly large. Even more daunting than the number of mailboxes, however, is the enormous size of the mailboxes—often more than 5GB each.
The sheer size of the mailboxes isnt surprising when you take into account the fact that many cases can take as long as seven years to resolve, with mountains of data shuffling among members of legal teams in the meantime.
Up until a few years ago, this degree of long-term message retention would have been unusual for the vast majority of companies. As a result of new compliance rules including the Sarbanes-Oxley Act, SEC 17a-4 and NASD 3010/3110, however, long-term data retention has become a business requirement in many industries, making e-mail storage management products such as SAFE and e-mail archive products such as Symantec Corp.s Enterprise Vault attractive solutions.
Indeed, with e-mail databases growing up to 200GB in size, something had to be done to keep Perkins Coies stores manageable and recoverable.
Enter EMC: “Our No. 1 reason for using SAFE was to trim down the size of our databases,” Ryan said.
Before splitting their overgrown databases into multiple stores of more reasonable size—to 20GB to 25GB each—IT staffers found that common tasks such as defragmentation required almost a full days worth of downtime to accomplish. Furthermore, the IT staff struggled to perform backup and restore operations in a timely fashion.
SAFE 2.1 monitors Exchange storage groups to make sure that storage is automatically allocated when it is needed, making day-to-day management of Exchange storage much easier. Without the monitoring and automatic storage allocation provided by SAFE, Ryan and his staff would have had an extremely hard time managing the dozens of e-mail stores in their environment.
None of this would have mattered, though, had SAFE not mitigated some of the complexities of SAN management. Of Ryans staff of 60 IT professionals, only four or five have extensive SAN training, so they appreciated SAFEs ease of use and tight integration, said Ryan.
Four Hewlett-Packard Co. ProLiant DL380 servers, each loaded with twin Intel Corp. Xeon 3.4GHz processors and 4GB of RAM, are used to run Perkins Coies Microsoft Exchange 2003 e-mail services. Each of the DL380 servers has twin Fibre Channel HBAs (host bus adapters) to provide dual, redundant paths to the SAN.
SAFE has also helped Ryan provide better uptime to his users. Ryan and his staff decided to use SAFE to keep e-mail up and running, as opposed to using a more traditional clustering solution. “Our attorneys are very demanding, and since they work round-the-clock, unplanned downtime is unacceptable,” he said.
With SAFE in place, Ryan and his staff members can quickly migrate a mail store from a fallen Exchange server to another mail server to ensure that attorneys and other staff can always access their in-boxes. “SAFE has worked flawlessly for us,” said Ryan. “We have had no downtime over the last 14 months.”
To further leverage Perkins Coies SAN investment, Ryan and his staff will be implementing a disaster recovery plan to establish replication between mail servers in the companys primary data center and its offices in Chicago. A dedicated 100M-bps link between the sites will handle replication traffic, according to Ryan.
With this plan in place, Ryan and his staff are hoping to keep recovery times to less than 1 hour, ensuring that users will be able to efficiently access e-mail from the Chicago servers in the event of a disaster at the data center.
Senior Analyst Henry Baltazar can be reached at henry_baltazar@ziffdavis.com.