Trouble there, is some of them at least know they need highly skilled talent, but it just isn't available.
Your point is well taken David, and pretty much acknowledged by the industry. What many companies fail to realize though is that when they do find or develop the talent, some earth shaking organizational challenges and decisions lay ahead.
From Gartner via Tech Republic.
Storage area management (SAM): Organization implications -
techrepublic.com
By Robert Passmore
Storage area management (SAM) implies change in organization and job descriptions. To achieve best practice, managers should treat the people and organizational aspects of SAM implementation with equal weight to the technical.
SAM helps suggest a change in thinking - that storage and data will be managed centrally by storage domain rather than by individual server/application. It follows that organization and job descriptions will change and that a new set of best practices will develop. Success is more likely if these changes are understood and planned upfront, rather than backed into as an afterthought to building a storage network.
Decision influence
When storage was attached to the server, storage decisions involved relatively few people. Storage networking, however, consolidates storage across groups of servers and applications, and perhaps across the whole enterprise. The result is that more people have an interest in decisions. These same people have valuable information about the storage requirements of applications and servers, about the company's networks, about current management applications and practices, and so on. How should the organization deal with this interest?
We have seen many companies in effect form a committee. This can be useful for gaining insight into the broader challenges and goals of a proposed SAM project, but if the committee becomes the principal decision-making body, or worse yet, a project implementation team, problems inevitably occur. (We won't go into the long history of why committees are what they are.)
The right approach is to view this group as an important source of information on the current environment, what is required, and what the goals should be for a SAM project. To maintain their interest and support, frequent communication about that project is important. Finally, this group may be a pool of potential candidates for the new work, but under a different job description.
Users should recognize the knowledge and experience of the existing organization and set up processes to formally mine the information, and to gain ongoing support of the group. However, new job descriptions and realignment of existing jobs are important parts of clarifying each individual's role.
Administrators and consumers
The most fundamental permanent organization change implied by SAM is the creation of storage administrators and storage consumers. This is not a question of whether to organize this way - it is a fallout of creating SAM. The SAM administrators plan, monitor, repair, report, and otherwise take care of the day-to-day running of the storage, the network, and the appropriate management functions.
The servers, applications, and related people have not disappeared, but rather have become consumers of the storage now managed centrally on their behalf. Consumers have needs, will make requests, and will want to know status. Administrators must be responsive to consumers, because the storage network doesn't exist on its own, but rather is there to provide a more efficient and flexible way to meet the needs of the applications. A good SAM architecture will make clear the options available to consumers and the responsiveness required of the administrators, and it will monitor service-level agreements negotiated between the two.
Users should design organizations and job descriptions with a full understanding of the administrator and consumer relationship. Goals, metrics, and reward systems should support overall business goals and the SAM architecture.
Storage utility
A number of vendors have described a "storage utility" as a part of their long-term visions for storage. This utility vision most typically describes storage as available from a socket in the wall, and delivered much as electricity, telephones, cable, or the Internet is delivered today. This is a consumer view where protection, backup/restore, performance and migration issues, and billing are all solved "magically" on the other side of the wall. (The utility term has also been borrowed by several opportunistic marketing groups to describe a much simpler storage or capacity-on-demand service.)
This consumer-oriented utility view can help organizations explain the long-term goals of SAM, and it can help motivate consumers and administrators to think in the new model. For administrators, the question is, "What do I need to do to act like the phone or power company to my consumers?" For the consumers, the question is, "What do I expect from a public utility providing storage services?"
Users should adopt and develop a utility-oriented vision to support the consumer/administrator relationship implied by SAM. The vision should be formally documented, and can be expected to evolve over time. Whether or not consumers actually are charged for the service, the administrators should behave like a utility and, at minimum, generate the usage reports and other information required to judge the effectiveness of the utility.
Skills
A frequently heard question is, "Should we train our networking people on storage, or our storage people on networks?" The answer is probably both, but with other skills required. One thing is clear: There are very few people who come already experienced in the wide range of skills ideal for SAM/storage area network (SAN) environments. This suggests a team approach with a mix of skills, and it also suggests that training is required.
The most important and difficult-to-acquire skills have to do with storage management. These skills are often hidden within the system-management ranks of today's organizations, although those who have implemented and operate centralized LAN-based backup and restore will have core elements of the skills. Networking knowledge is important, although SAN topologies are small relative to WAN operations today. As SANs grow, and particularly as they bridge from site to site over WAN technology, networking skills will become more important. Knowledge of the OS, file system, and database for each application deployed is critical for backup/restore, volume sizing, and performance-tuning activities. Planning, financial, program management, asset management, and other general management skills are needed. Finally, in an administrator/consumer relationship, the people interface and service skills are important.
Users should recruit a carefully chosen set of skills to implement SAM and be prepared for extensive training over time.
Training
Users report difficulty finding independent training. Courses are available on detailed technical subjects like the Fibre Channel (FC) protocol but not at higher levels. The most accessible events are vendor seminars, which provide helpful overviews but often are part of the pre-sales process, and are biased to the vendor and gloss over the challenges. Once a vendor is selected, technical courses designed for the vendor?s own field engineers can provide in-depth installation and operation information, but reports indicate that these are usually component-oriented and fail to cover the higher-level integration. User feedback on nonvendor training companies indicates limited offerings and narrow topics. Universities offer little about storage in general, much less on specific topics like FC or storage resource management.
On the other hand, training for IP-based networking is much easier to find due to IP?s long history and broad adoption.
Conferences and vendor-supplied courses are the best resource today. Users should consider customer training as a part of vendors? deliverables in an overall SAM/SAN deployment. But make sure that the vendor?s professional service organization has the skills to provide the necessary training.
Strategic planning assumptions
- Most, if not all, workers currently involved in some aspect of storage management will see new jobs or changes in their job descriptions as SAM projects move to best practices within the organization.
- Storage management software, fully SAM-aware and with the architecture to support the administrator/consumer model, will not be generally available until at least 2004.
- The most successful SAM deployments in large organizations will be those that implement a formal utility model for storage.
- The shortage of people trained and experienced in storage management and other related storage engineering skills will continue until at least 2004.
- Lack of adequate training will remain a major inhibitor of successful SAM/SAN deployment until at least 2004.
Bottom line
To achieve best practice, managers should treat the people and organizational aspects of SAM implementation with equal weight to the technical. Mine the knowledge and ideas of existing staff and communicate to build support. Organize with recognition of the administrator/consumer relationship. Choose staff for the necessary skills, but be prepared to train. Finally, drive to a storage utility model as a goal.
Gartner originally published this report on May 21, 2001. |
