Vendor Accent

Storage economics: The changing paradigm

ROI and reduction in long-term operational expenses should dictate storage investments and expansion, says Srikant Chakrapani

There is growing realisation that storage economics is much more than determining price-per-megabyte metrics or total cost of storage ownership. Today, storage economics takes a broader view as to where costs may lie within the storage ecosystem, how to extract these costs, and how to think strategically rather than tactically regarding storage deployment and architectural decisions. Purchasing low-cost storage solutions does not equate with lowering operating expenses or reducing the total cost of storage ownership. Purchasing cheaper disk solutions to reduce short-term capital expenses (CAPEX) can result in a negative long-term impact on operational expenses (OPEX). Moreover, storage-related prices demonstrate inelastic trends relative to demand. In other words, vendors lowering their prices on unit costs of storage will tend not to increase demand for the product. This is because storage demand is driven by applications, business growth, data exploitation and compliance laws—and not necessarily by lower prices.

The correct approach for storage economics decision-making (and an industry best practice) is to take a strategic view of storage architectures and determine the best solutions based on TCO and OPEX minimisation.

TCO, ROI: the right perspective

A proper examination of storage economics requires consideration of two perspectives: total cost of operational ownership (TCO) and return on investment (ROI). The perspectives of TCO and ROI are similar in computation and development, but fundamentally different in the types of decisions for which they should be used.

First, ROI is an effective method to use when proposing a new set of actions or activities to replace existing practices or technology. Each of the investment options has to be defined as to its respective return on investment. Basic ROI results have to answer three questions: How much is the investment, how fast is its payback, and how much is the total or net savings on this investment.

ROI is crucial for demonstrating to management that even with enough incremental disk capacity to meet the basic company needs, the change to a new architecture or storage solution can provide positive ROI due to the OPEX reduction in many cost-sensitive areas. In the case of planned storage growth or expansion (reactive in nature relative to demand), TCO can be effective for calculating total lifecycle costs of competitive or comparative solutions. Total cost of ownership is exactly what the name implies—the total operating and purchase cost of an asset (such as storage). To determine TCO, several costs that have accumulated over a number of years need to be taken into consideration. These include elements purchased from the vendor (hardware, software, installation, verification, migration); IT departmental expenses for installation (new training, new room preparation); write-off costs if the systems being replaced are not fully depreciated or are not at the end of the lease life; year-on-year environmental costs such as electricity, air-conditioning and floor space; and maintenance costs (after the warranty period) for hardware and software.

OPEX is for real

Before any actions or calculations are attempted, it is worth determining how much storage OPEX potential exists, and what efforts are required to discover, harvest and reduce the actual storage costs. On average, for every 12 TB of usable disk capacity within the storage infrastructure, there is $1 million net OPEX savings potential.

It is a significant finding to be able to generalise or predict the amount of OPEX money that currently lies within a storage infrastructure. Many situations generate $1 million of OPEX savings with five or six terabytes of total storage, while some environments generate smaller savings. The $1 million to 12 TB ratio is conservative, and is generated over a three-year term.

OPEX reduction: some options

The OPEX reduction strategy can be defined on the basis of OPEX reduction potential and effort, investment, time and labour needed to accomplish these activities. For example, if space, electricity, hardware maintenance and lower labour costs are desired, then one of the primary activities should be storage consolidation.

The first level includes OPEX cost-reduction activities that are most easily accomplished with proven results and demonstrable savings. These are the foundation activities that one should start with before moving to more advanced initiatives. Such actions pick off low-hanging fruit, and tend to be where most of the $1 million per 12 TB savings reside. These activities should be considered and acted upon first in any OPEX-reduction strategy. Some of the activities at this level include storage consolidation, storage networks, common backup, hierarchical storage management and storage management automation.

The second or middle level includes projects and initiatives that go beyond basic infrastructure investment activities that work to optimise processes, procedures, organisational alignment and so forth. These efforts are common to (or scheduled by) more mature IT departments. This level may be seen as enabling more advanced strategies and OPEX-reduction strategies. For many IT departments, this effort is 12-18 months out, or after the activities in the first level have been achieved. Many of the savings from this area are considered to be soft savings, but IT organisations with best-in-class storage operations (and costs) have implemented and mastered these functions. Activities in this level include best practices and processes like change control, version control, capacity plans, storage chargeback, storage resource management, centralised storage management, multi-tiered storage options and building and managing storage architectures.

The third level includes actions that are more distant in the future. Many IT planners are targeting these activities as the end-goal (storage utility, for example). These actions may require a fundamental change in business methods or business or IT roles for implementation. New technologies and storage developments will tend to appear in this layer first, when experience proves them to pass the OPEX test of time. Activities in this level include storage utility, virtualisation and storage provisioning.

Conducting an economic analysis

After realising the real OPEX money potential that exists in an organisation, any further motivation to uncover OPEX potential must come from the IT or finance department. There are certain situations or trigger events that may help define conditions or situations where OPEX potential exists, and where an in-depth economic analysis may be in order. If some (or many) of these trigger conditions apply, consider undertaking an economic analysis of the enterprise’s storage environment. There could be enough OPEX potential to warrant a change in vendors, topologies, protocols, operational practices or deployment. For example, IT organisations having a lot of direct-attached storage spread through the enterprise, or organisations where storage outages exist, can provide a suitable trigger for undertaking economic analysis. Some of the other trigger events include places where IT plans and strategy like application and server consolidation; presence of multiple RAID storage systems in the data centre (small-to-medium systems); the need for IT departments to reduce operating expenses; and a situation where finance and IT work together leading to greater involvement of finance in IT decisions.

The author is Consulting Director, Hitachi Data Systems