By Russ Fellows
Transitioning into a new budget cycle is an opportunity to review your IT strategy and one of the biggest components of IT — storage. This article is designed to help IT managers and administrators understand how to prioritize potential projects by ensuring they are aligned with business goals, which ultimately results in the best storage decisions.
Business executives have set a few directives for IT departments. They can be summed up as: "Lower costs without lowering results." In addition, executives are looking to move IT spending from capital to operational expenses, which can improve flexibility and corporate effectiveness.
As a result, IT departments are challenged to improve IT efficiency (do more with less), reduce risk (disaster recovery, compliance and related issues), and move Capex to Opex (change from purchasing IT to using IT as a service).
There are a myriad of new technologies vying for attention. Each new technology typically is billed as ‘the next best thing' that will solve the problems companies are experiencing. However, IT staff and business executives alike know that new technologies are often sold as stand-alone products, with little integration to other technologies.
Integration with the existing IT infrastructure should be a priority, as opposed to stand-alone fixes, which can increase complexity and total costs. Proposed solutions are impractical unless they integrate with existing infrastructure. Projects should add incremental changes, with each project able to show tangible results.
Listed below are some of the technologies IT departments are implementing. Some have been around for a few years and others are just emerging.
- Virtualization (server and storage)
- Automated tiering (ILM gets its day)
- Thin provisioning
- Data deduplication
- IT as a service (cloud-based IT, including infrastructure, platforms and software as services)
- Converged Ethernet SAN infrastructure, and Fibre Channel over Ethernet (FCoE)
- Solid-state disk (SSD) drives
- Private and/or public clouds
This list is not exhaustive, but it includes the type of projects many users have recently started, or are looking to implement in the near term. What is needed is a method for prioritizing investment choices that help your company meet its goals while ensuring IT is recognized for the value it brings to the organization.
Aligning business with technology
One way to view potential storage projects is through the filter of how they align with top-level business objectives. There are only two options available that can shift capital expenses to operational expenses: outsource IT or use IT as a Service (ITaaS). Neither option is likely to reduce risk, or improve service levels, although both can improve efficiency.
For IT managers and other executives evaluating potential IT projects, the following table should help to clarify what areas each type of project may impact. In order to quantify the improvement, it is necessary to evaluate each project and the unique organizational factors.
Efforts that seek to improve IT efficiency are top of mind for many businesses in 2010. Efficiency improvement will typically mean better results and lower cost. The most successful projects will be able to demonstrate that they both improve results and lower the cost of IT. The criterion for IT project approval is often a three-year ROI, which corresponds to the typical three-year equipment life span. More recently, the three-year period aligns with budget planning horizons as well.
In general, risk reduction is very important to IT departments. One area where this is seen is in the area of data protection, including backup and replication. In times of economic difficulty, there is a temptation to delay IT projects that reduce risk, since the benefits have an indirect effect on revenue. However, the benefits of risk reduction should be quantified so that they can be measured accurately.
Improve service levels
Service level improvement projects are likely to take a back seat to other projects in the next year or two. The exception to this will be business applications that can demonstrate that higher service levels will in turn drive higher revenue. Projects with the objective to improve internal users' service levels should not be the primary focus, particularly as IT organizations look to demonstrate that they are aligned with company objectives.
Best investments for 2010
The importance of data protection to organizations can never be under estimated. Not only can data protection help reduce risk, it can improve productivity and efficiency. Typically, human events are the most common reason for loss of information. In order to protect against this type of data loss, it is cost effective to use point-in-time (PIT) snapshots to make copies of data.
Thus, projects related to data protection should be given a high priority, particularly if investments have not been made recently. Projects can include those that reduce risk, such as adding off-site disaster recovery capability, or those that help improve service levels. In terms of disaster recovery, the first line of defense should always be local PIT copies of data on disk. Point-in-time copies can be created and managed by the operating system or storage system. In most cases, the cost is low, even when considering the additional storage.
Another area of focus in the area of backup and recovery is the use of disk rather than tape. Companies not using disk for backups —either disk-to-disk (D2D) or a virtual tape library (VTL) — should consider this as an area for investment. Disk should be a part of every IT organization's data protection scheme. In general, disk should always be the first line of defense in protecting against loss of data from error or corruption. Tape (when used) should be used as the last line of defense, offering protection for disaster recovery and long-term archive.
Another area of IT investment is storage optimization. Two popular optimization technologies include thin provisioning and data deduplication. Both of these technologies can have immediate payback if implemented correctly.
Other storage optimization technologies may require the use of virtualization or data migration tools. The use of solid-state disk (SSD) drives, archiving and storage tiering all require the ability to transparently move or migrate data in order to see results. Thus, these storage optimization projects should be delayed until storage virtualization and/or migration tools are in place. Typically, data migration tools are available, although they must be virtualization aware in order to be effective.
Investment in SSD technology can improve efficiency, due to a reduction in the total number of storage devices required. Although the cost of SSDs remains high, the cost per I/O is often less than that of traditional disks. As the need for I/O becomes more important, and as the prices of SSDs drop, more companies will find it cost effective to deploy SSDs.
The premise of cloud computing is that IT can be consumed as a service. This implies little to no capital investments. For smaller companies comfortable with consuming IT services externally, this vision can be a reality. Thus, for companies with limited budgets or IT infrastructure, ITaaS or cloud computing/storage can be practical.
For the vast remainder of companies, the term "cloud" is used as an attempt to sell equipment to IT departments. Thus, for these companies, investments in cloud computing are likely not the best place to invest their critical IT expenditures over the next two to three years. A significant amount of technology improvements must be made in order for cloud or ITaaS to be ready for widespread enterprise deployments.
There is no single answer to the question of what type of storage project will deliver the best results. The answer is highly dependent on the size of the organization, the type and amount of existing equipment, and the goals. However, by clearly understanding what the business objectives and priorities are, it is possible to create a prioritized list of projects that will align with those objectives.
Ultimately, business and IT executives will view projects as being successful if they are aligned with the goals of the organization. Demonstrating success with projects in business terms will help demonstrate the alignment. A multi-year approach will improve chances of achieving a ROI.
Small IT organizations
Companies that are looking for new storage projects that total less than $100,000 are the most numerous. For these smaller organizations, the focus should be on projects that improve the backup process or those that simplify and consolidate storage platforms. Virtualization projects are typically too complex for smaller companies, so migrating to new storage platforms is the most effective form of consolidation.
Smaller firms should also leverage the capability of IT-as-a-Service offerings. Possible projects include moving email, backup or other services into the cloud. These efforts can significantly reduce the total cost of IT for small companies, and may improve overall service and availability levels.
For companies looking to invest $100,000 to $1 million over the next year for new storage, there are several areas of focus. Implementing D2D, VTL or other disk-based backup technologies can reduce risk and improve performance. These products often support data deduplication, which improves efficiency.
Other projects that are successful in mid-size firms include thin provisioning and storage replication for disaster recovery. Other choices may be difficult to implement within the available budget.
Companies that will be spending more than $1 million in the coming year on new IT storage projects should have a somewhat different perspective. Organizations of this size will often have multiple types of storage systems and generations of systems in place. Consolidation and tiering are two projects that can provide benefits. In order to reduce the impact from these operations, some type of virtualization technology is also helpful to facilitate these and other projects.
In addition, the heavy use of virtual server technology has resulted in a dramatic rise in the number of system images. Each virtual server requires storage, which is often highly redundant. Thin provisioning and data deduplication can both significantly reduce the amount of overhead associated with virtual server installations.
Finally, for very large companies or those looking to become service providers, IT as a service and cloud computing projects may be viable. However, considerable domain expertise and integration will be required.
RUSS FELLOWS is a senior partner with the Evaluator Group research and consulting firm, www.evaluatorgroup.com.