Network Computing is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

A Compelling SAN For IT: Page 2 of 3

MORE THAN TIERS

Another well-implemented feature is the replication engine's bandwidth throttling and quality-of-service features. While all of the major players offer simple snapshot and replication functionality, what if you need to replicate your data over a WAN link? Storage Center allows you to define a replication schedule that's sensitive to bandwidth.

As a result, admins can replicate critical data off-site and have complete control of when to transfer data, and how much of the WAN link to use for replication. The drawback is that you'll need to replicate with another Compellent SAN to use this functionality.

Compellent Storage
Center 30 Series
Tiered storage Tier 1 15,000-rpm FC drives;
Tier 2 10,000-rpm FC drives;
Tier 3 7,000-rpm SATA drives
Slot configuration 5 PCI-E slots, 1 PCI-X slot
FastTrack Moves frequently accessed data to outer edge of spindle to improve seek time and disk performance
Dynamic Block Architecture Can arrange data movement and access at the block level
High availability Hot-swappable and redundant power supplies, cooling fans, and controllers

We were also quite impressed with the way Compellent executed firmware updates. Upgrading code on a SAN is a stressful proposition and usually means downtime while the code is applied. With Storage Center 4.0, you can download and prepare the code upgrade for installation while the system is still running. A three-minute reboot of the array then completes the promotion of the new code into production. For fault tolerance, the firmware is both kept on a standby flash card and striped across the drives in your SAN, so you can easily revert to old code if a problem arises.

One negative we noticed was with respect to the reporting of critical events, such as a failed disk, which is done via TCP/IP. That's not a problem, as long as the network is functioning normally.

In addition, we don't particularly enjoy providing bidirectional communication through our firewall to a critical resource. EMC, by contrast, typically attaches a modem to a private analog line for alerts and remote troubleshooting.