Interesting all-cash acquisition, J. McData booked $4.8M in software and services revenue last quarter based on its E/OS (device management) real-time operating system and EFCM (storage network management) software. Because of its history, Brocade is going to have difficult time competing with McData's leadership in providing heterogeneous SAN support which is going to be enhanced by this acquisition.
According to Soundview, Brocade had 2,000 end users at the end of the July 2001 quarter. That compares with the 900+ end users that McData disclosed that it had at the end of the December 2000 quarter.
Of Brocade's 2,000 end users, the top 500 customers averaged 300 ports with the top 50 customers averaging 1,500 ports.
BROCADE's CUSTOMER PROFILE
Source: Soundview
Class Average
TOP 50 Customers 1,500 ports
NEXT 450 Customers 167 ports
BOTTOM 1,500 Customers ?
Brocade claims to have shipped more than 900,000 FC ports. If the 500 largest customers average 300 ports, that accounts for 150,000 ports leaving 1,500 customers with 750,000 ports or an average of 500 ports! The numbers don't add up even if you use the McData estimate of 600,000 total FC ports shipped!
McData hasn't release comparable numbers, but most of the early adopters of director-based SANs came from the ESCON SAN environment, typically customers migrating secondary applications from the mainframe to high-end Unix. McData's installed base for ESCON SANs was around 5,100 sites in more than 65 countries. This is why McData had the largest SAN in 1999 (128 ports) and 1H2001 (6,000 ports).
Anyway, here's a fairly recent article penned by a SANavigator engineer.
The case for SAN SRM applications
The goal is to reduce storage management costs in heterogeneous storage network environments while improving overall return on investment.
BY STEVE HINDMAN
Advances in storage technology have led to a critical need for Storage Resource Management (SRM) software. Forrester Research reports that by 2003, a global 2500 company will add an average of 150TB. By comparison, that same company added an average of 15TB in 1999. To merely guarantee critical storage metrics is not enough. The ability to exchange information immediately-dubbed "zero latency" by Gartner Group-across multiple platforms is necessary to remain competitive.
Promising improved scalability, availability, and data protection, storage area networks (SANs) have emerged as a key to meeting data storage requirements. As storage expansion gains momentum and the infrastructure evolves, storage management becomes a primary challenge.
Keeping pace with problem diagnosis is increasingly difficult as the propagation of cascading events rapidly overwhelms multiple applications. As a result, management challenges are proportionate with escalating SAN infrastructure complexity.
Simple one-switch SANs are easily managed by the switch vendor's application. However, as a SAN is expanded, a variety of vendors' devices are introduced, each requiring another management application.
The task of monitoring and managing becomes an arduous responsibility. The storage manager jumps from one application to another, entering and inspecting duplicate information. This increasing complexity and the resultant drain on management resources are threatening the primary promise of SANs: decreased storage management costs.
A switch manufacturer's proprietary management application is incapable of addressing a heterogeneous network. The highly competitive nature of maturing technologies discourages cooperation between manufacturers. Storage management standards are perpetually evolving, with each manufacturer adopting new features before and after a standard is introduced.
Storage Resource Management
In order for SANs to deliver on their promises, a single centralized SRM application is required to perform the tasks of managing the complete storage environment. Key SRM tasks include discovery and visualization, device management, and performance and status monitoring.
Discovery and visualization
SRM applications automate one of management's most tedious tasks: visualizing the storage environment. Previously requiring hours of manual diagramming, a visual map of the network can be prepared in minutes.
Visualization begins with the automatic discovery of all devices in the storage environment, which saves time in the initial discovery and whenever new devices are added. The SRM application recognizes each device and how it should be managed and monitored and then assists in its integration into the storage enterprise. The information gained from the discovery process is saved so that changes may be recognized-a feature called "persistence."
Using the information from the discovery process, a visual map of the storage environment is prepared, presenting all devices and their interconnections. This visualization highlights redundancies within the SAN, allowing management to visually confirm high-availability designs.
Device management
SRM applications can significantly simplify device management either by directly configuring each device or, where necessary, by automatically launching the appropriate device-specific application.
Some of the most difficult tasks are best performed directly within the SRM application. Zoning - the process of virtually separating a SAN into isolated zones that are protected from interfering with each other-is performed differently by each switch and router manufacturer. A single interface to all devices can simplify this task with an intuitive visual display.
Performance and status monitoring
A failure within the storage environment can affect every connected application. For example, a failure in a SAN with redundancy may not be immediately noticed yet may cause SAN performance to degrade significantly. For this reason, the SAN and all storage devices must be continuously monitored in an automated fashion.
SRM applications continuously monitor every device in the storage environment. All events are logged and addressed as predefined, which may include paging IT management. Events range from minor warnings such as a cooling fan RPM drop or a redundant power supply failure to major events such as a severed cable or device failure. The SRM application will further poll each device on a regular basis, ensuring the loss of a device or connection is promptly reported. Logs provide the critical information for diagnosing problems or performance degradations. In addition to device monitoring, performance and capacity information may also be tracked. User-specified thresholds allow SRM applications to notify administrators when certain conditions exist.
Beyond monitoring storage metrics, SRM applications typically collect reporting data to facilitate dynamic trend analysis. This information is critical to optimize performance and forecast capacity and consumption requirements. This analysis and reporting allows management to proactively address storage demands.
SRM applications are critical to a highly available, consistently performing storage environment. Although it is possible to establish and operate a large storage enterprise or SAN without a global management application, it is a significantly greater task requiring more IT management resources, which in turn reduces the return on investment of the SAN. Implementing an SRM application can substantially simplify the task of storage management, reduce overall storage costs, increase data availability, and diminish the potential for error introduced by configuring individual devices using independent, proprietary applications.
Steve Hindman is senior software product manager at SANavigator (www.sanavigator.com) in San Jose, CA.
InfoStor June, 2001
is.pennnet.com |
