Tip

Reap the benefits of intelligent storage

Key criteria for an intelligent operating environment

To enhance the long-term performance, availability, manageability and cost-effectiveness of their storage services, customers should analyze the capabilities of operating environments in five key areas: Parallelism, prioritization, integrity, application support and maturity.

Parallelism is the ability to handle simultaneous I/O requests from upwards of thousands of devices, ranging from servers to switches in a network storage fabric. This is particularly important when several different applications, or several different users, need access to the same data at the same time. In high-end transaction processing systems, parallelism can allow the same application to access multiple disk drives to maintain response times as the number of users spikes unpredictably. In high-end scientific applications, such as DNA sequencing, parallelism can allow multiple applications to access the same data to speed research.

Mid-range applications and storage systems can also take advantage of parallelism. For example, a mid-range storage array

Requires Free Membership to View

running an ERP application might use multi-path I/O capabilities to simultaneously operate multiple ports to a host server. Whatever the range of capabilities needed, parallelism allows the storage devices to handle new applications, new users and new types of storage while providing the agreed-upon levels of service to customers.

Prioritization is the ability to look at all the incoming requests for storage services and decide, on the fly, which should be handled when and by which physical or logical storage resource. The ultimate aim is to "handle the most important I/O first" so that service levels are met despite changes in the IT infrastructure. At the high end, consider a financial services company running a new decision-support application that analyzes the profitability of various bank loans. Because the new application scans multiple databases, it will read and write data to storage devices far differently than an existing application that processes one loan at a time.

At the mid-range level, prioritization might mean the ability for a customer to hot-swap a failed disk or fan without interrupting application access to the data. However it is accomplished, a storage operating environment can deliver cost-effective, high levels of service amid changing conditions.

End-to-end integrity plays an important role. The storage operating environment must continually assure the reliability of the hardware, software and application data. As hardware is monitored, potential problem areas must be identified, escalated and recovered in all circumstances. The software environment itself must repeatedly check key data structures and control information for corruption. And finally, the operating environment must make sure that application data integrity is not compromised on its journey from application to disk and back again.

Storage applications are rarely seen by users, but are instead infrastructure-level, mission-critical applications that run in the data center. These include backup to preserve data, mirroring to keep critical applications running and dynamic storage provisioning which can reallocate storage resources without slowing application performance. Powerful storage operating environments ensure that all applications work at peak performance while not stepping on each other and not impacting host service levels.

The maturity of the storage operating environment means the software which makes up the operating environment is reliable, scalable, well-tested and well-supported, all crucial considerations for customers who are planning and managing for the long-term. Operating environments are highly complex and require a significant amount of seasoning and maturity before they can be trusted to handle the highest service level conditions. Most organizations would hesitate to run a mission-critical application on a new operating environment, database or operating system. The result would likely be more than an application crash and could have serious implications on the entire storage array. The same holds true for storage operating environments. The vendor providing a mature operating environment also has rigorous quality control and testing processes, and proven change management and version control processes to help customers add new capabilities over time.

Look back over the history of the computer industry and one thing becomes apparent: Over time, software rises in importance when compared to hardware. No one buys a PC no matter what its specifications if it can't run his or her favorite applications. No IT manager considers looking at server hardware until he decides which applications he needs to run, on which operating system.

The same evolution is now happening in storage, where the equivalent of the operating system is the operating environment: The device-level intelligence that sits between the physical storage and the rest of the IT infrastructure. Traditionally referred to as "microcode" and originally designed to handle only the most routine chores such as data correction and caching, a range of intelligent storage operating environments has emerged. Optimized for various application needs, storage operating environments will play an increasingly strategic role in delivering the most advanced and cost-effective storage services over the next several years.

Customers who choose storage hardware with the appropriate level of intelligent operating environments will achieve greater performance, reliability -- and the ability to exploit new capabilities over time -- at much lower costs than those who choose storage hardware purely based on raw "speeds and feeds." This article will define intelligent operating environments, the increasingly important business benefits they will deliver, and the key features customers should look for in such environments.

Hardware vs. software: A lesson from history
Eventually, all computing devices become more intelligent. Whether cell phones, PDAs or DVD players, hardware manufacturers perform the annual ritual of cramming more raw intelligence and storage capacity into smaller physical devices at lower cost.

But the raw capabilities of the hardware, impressive as they are, pale in importance to the software that runs the hardware. Over time, software (and specifically application software) drives the choice of hardware.

When evaluating server hardware, for instance, IT managers don't first choose the microprocessor inside the box or the hardware vendor who makes the box. They first choose which operating system (such as Windows, Unix or z/OS) will run the applications they need, cost least to manage and maintain, and be backed by a vendor who will provide quality support and a smooth migration path to new technologies.

In some respects, today's storage market resembles the server market in the late 1980s, when the arcane depths of rival architectures for microprocessors and motherboards received disproportionate amounts of attention. Within three years, most storage customers will be thinking less about chips and specs and more about strategic software capabilities made possible through the intelligent operating environment.

Leading-edge customers are already demanding key capabilities in the operating environments that ship with their storage subsystems. There are five capabilities, each of which has varying importance depending on the customer's size, types of applications they run and other requirements. They include the ability to:

  • Provide massive parallelism, enabling a multitude of activity to take place concurrently.

  • Ensure the prioritization of tasks and overall adaptability, so performance levels are met regardless of circumstance and the system adapts easily to changes in workloads, hardware failures, etc.

  • Assure the end-to-end integrity of the hardware, software and application environment.

  • Offer access to the broadest range of powerful storage applications to protect, manage and share critical corporate information.

  • Demonstrate a proven track record of maturity in the architecture's stability and robustness.

IT operations that are the first to realize the strategic importance of choosing the right intelligent storage operating system will be first to achieve higher performance and uptime, reduced hardware purchases through more efficient use of existing storage and reduced long-term management costs.

The intelligent operating environment
A storage operating environment is, roughly speaking, the equivalent of the operating system in a PC or a server. It controls the flow of data to and from the physical storage device, as well as within the device (such as among the disks in a RAID array.) Because it usually takes the form of firmware embedded in the storage device, it is often misunderstood as a low-level piece of the storage "plumbing" that handles only routine data-handling chores.

But as storage devices have grown more complicated and capable, so have their operating environments. In some storage systems, the "microcode" has grown into adulthood with upwards of two million lines of code, and is run by microprocessors more powerful than the hosts to which the storage is attached.

It's no accident the storage operating environment is so important, because it sits right between the data which is the lifeblood of any enterprise and any application, device or user who needs that data. Rather than mere passive "plumbing," the proper intelligent storage operating environment is a crucial tool in making the most effective use of storage resources, driving down storage management costs and delivering superior, cost-effective storage services.

In summary
Intelligent storage operating environments are far from commodity "microcode" that ships with the disk. Within several years, they will become strategic enablers of quality storage services. Customers who begin choosing storage hardware now based on the capabilities of its operating environment (such as parallelism, prioritization, integrity, application support and mature development practices) will be the first to be rewarded with significant and long-lasting cost/performance benefits.


About the author: Chuck Hollis, EMC's vice president of platform marketing, is responsible for setting market strategy for EMC's platform offerings across all major market segments. Prior to this role, Hollis led EMC's European Professional Services and Systems Engineering organizations.

Hollis joined EMC in 1994 as part of an initial team to help the company enter the open systems storage market. Prior to joining EMC, he held various management positions at several UNIX and database companies in Silicon Valley. He has degrees in Computer Science and Economics from the University of California at Santa Cruz.

This was first published in January 2003

There are Comments. Add yours.

 
TIP: Want to include a code block in your comment? Use <pre> or <code> tags around the desired text. Ex: <code>insert code</code>

REGISTER or login:

Forgot Password?
By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy
Sort by: OldestNewest

Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to:

Disclaimer: Our Tips Exchange is a forum for you to share technical advice and expertise with your peers and to learn from other enterprise IT professionals. TechTarget provides the infrastructure to facilitate this sharing of information. However, we cannot guarantee the accuracy or validity of the material submitted. You agree that your use of the Ask The Expert services and your reliance on any questions, answers, information or other materials received through this Web site is at your own risk.