Storage Magazine - UK
 

THE CASE FOR SMARTER STORAGE

From STORAGE Magazine Vol 12, Issue 01 - February 2012

IS IT SENSIBLE TO BE AN EXTREME SPORTS ENTHUSIAST AND THEN GO WITHOUT INSURANCE? SO WHY WOULD YOU FAIL TO BACK UP THE VITAL INFORMATION OF YOUR BUSINESS? BRIAN WALL STRUGGLES TO FIND THE ANSWER!

Data continues to grow at a remarkable rate and clearly has no intentions of stopping. But this alone does not make an organisation more competitive. It is the way in which the data is used and manipulated that enables businesses to create unique competitive advantage. And to do this successfully requires the correct systems and storage for the business.

This data growth means organisations are having to buy more infrastructure that takes up more data centre space and incurs more cost. The other consequential impact is on management of the environment. "Multiple systems with differing management interfaces increase the complexity of the environment, which today could be being managed by yesterday's server administrators," points out Ian Shave, IBM UKI Storage brand leader. "The skills required for managing such an environment can be hard to find in the industry, even assuming there is operational budget to allow for additional staff."

ENERGY CONSUMPTION
Historically, storage densities were increasing and costs were declining faster than the data was growing, but technology growth is slowing, while data growth rates are accelerating. Other consequential impacts of installing the required additional infrastructure include those on energy consumption. With the more physical infrastructure put on the floor, the more disks need powering, the greater the energy consumption becomes.

"The actual data size, compared to what was allocated to a requirement, is also, in the majority of cases, significantly smaller," he continues. "Capacity projections for projects are frequently over specified, often requesting the projected capacity requirement for a three or more year period, as well as factoring in contingency in the event the project was undersized. With average utilisation of storage at 30%, organisations are getting a poor return on investment from a depreciating asset, as well as wasting energy consumption and associated operating costs.

"To help visualise the inefficiencies of this low utilisation, this is like having a 10-storey building, with only three floors occupied and then, when employing 20 more people, buying a new building to put them in. This is, in effect what many organisations are doing today: believing they need, and subsequently buying, more storage."

STORE MORE WITH WHAT'S ON THE FLOOR
Through storage virtualisation technologies, all usable capacity can be grouped into storage pools, enabling it to be carved up without wastage. Think of multiple partially filled glasses of water versus a jug of water being shared with straws, so there is only one resource to think of from the point of view of keeping it as near to full as possible, suggests Shave.

"Thin provisioning allows you to allocate more storage capacity than you have available - the concept of over-provisioning - so reducing that wastage associated with a project that has been over-specified. The efficiency gains from these technologies can be significant. By combining these technologies, it's possible that 'allocated' capacity is actually higher than the usable capacity."

DATA IN THE RIGHT PLACE
With the right storage management tools, any allocated capacity that is not being accessed can be clearly identified. If it is a volume that actually contains no data, it can be released and added back into the storage pool. If it contains data of value that is not being accessed, with the right automation tools linked to policies from the reporting, it can be moved to a more cost-efficient technology. The reporting tools also become highly valuable for those organisations that utilise thin provisioning technology, providing capacity forecasting and reporting, ensuring the right physical capacity is available to satisfy the now over-allocated storage.

"Automated tiering dynamically redistributes active data across multiple tiers of storage class, based on workload characteristics, enabling efficient use of flash storage to reduce response times, improve application performance and service," states Shave. "This identifies data that is inactive, or with a very low access frequency, to be stored on more cost-effective storage classes, lowering overall costs."

REDUCE DATA STORED
Finally, data reduction technologies, such as compression and deduplication, ensure data takes up less physical capacity, and with compression technologies that can be applied to production data without compromising performance, the benefit of these technologies becomes more accessible. IBM's Real Time Compression technology, for example, offers organisations an opportunity to transform their return on storage investments. For many organisations, the ability to dramatically increase the utilisation of existing assets could result in not needing to purchase more infrastructure for a significant period of time, taking much needed pressure off capex budgets, according to Shave.

"Storage virtualisation and automation technologies simplify the management of the environment and increase the efficiency, leading to higher data protection. Through the combination of all these technologies, organisations can reduce both costs and energy requirements, helping to reduce operational costs. Very importantly, these technologies and efficiencies can be applied to an organisation's existing infrastructure and help gain significant additional value from existing assets."

DATA PROTECTION AND RETENTION
The explosion of information is compounded for data protection and retention, because organisations have been keeping more data and storing it longer. Lack of understanding of legislation/compliance requirements and what data needs to be kept, and for what period of time, has led many organisations to the conclusion: keep everything for ever.

"The impact of this is that organisations often find themselves spending money in ways they'd rather not, spending nearly as much to protect and retain data as they do to support revenuegenerating applications. Typically, data is accessed most frequently early in its life after being generated. This can then tail off to a very low to zero access frequency. However, should that information be updated in any way or should data analytics be run on historical data, its access frequency is likely to again increase significantly, albeit potentially for a shorter time period. This activity will again tail off over time, eventually to zero. This process could happen multiple times before finally the data access stays at zero for an extended period of time.

If a data type is considered business critical, with a very high availability and performance requirement, while the data is being frequently accessed, it's possible we would want it on the highest performing disk system, but also replicated synchronously to another data centre."

LOWER, CHEAPER TIER
When its access frequency goes down, it would be more cost efficient to move it to a lower, cheaper tier of storage, he adds. "If it's updated, we again may want to move it back to a higher tier to match the likely higher access frequency of the data. However, during this whole process, not only does the data need to meet the performance level determined in the profiling, but its data protection or integrity must not be compromised." Separating data protection technologies and functionalities from the infrastructure helps to deliver a more costeffective infrastructure, but crucially, without compromising data protection.

Another common issue contributing to the inescapable data growth is multiple copies of the data; not only due to multiple copies across multiple applications or users, but also from storing on primary storage to snapshots, replicated to backup and archive, DR test etc. To dramatically reduce this impact, which is becoming compounded as data grows and data protection requirements increase, technologies such as data deduplication can help significantly reduce the number of copies actually being stored.

SMART ARCHIVING
"Deleting data is often so difficult to reason with, so having the right archiving strategy could be very financially beneficial. 'Smart Archiving' can help satisfy new use cases for data retrieval, improve the performance and effectiveness of production systems, and reduce the amount of data that has to be backed up. Of course, particularly if archiving data to tape technologies, this can help significantly towards the overall business requirement of reducing carbon footprint and reducing operational costs," he maintains.

"You don't get much more energy efficient than putting data on a tape cartridge. Once it's on the cartridge, it isn't consuming any energy at all. This has led to a resurgence in the tape market, as more companies realise their commitments to reducing energy consumption, along with the need to streamline costs."

In order to move to a 'Smarter Storage' system, an important first step is to reassess the existing infrastructure, he adds. "Priorities for data types, files types or applications could have changed since this was last assessed, but it's also possible they were measured to fit to pre-existing infrastructure inflexibilities, rather than truly evaluated on the requirements to meet the business objectives. The final step is to develop a transformation plan to transition the organisation from where they are today to how they would like things to look tomorrow."

Whether organisations have a focused path for their storage needs, or are not sure where to start, or indeed any situation in between, IBM believes it has the systems and capabilities to help organisations unlock their unique competitive advantage using smarter storage infrastructure. Certainly, the data explosion that touches everyone is only going to accelerate. Failing to tackle it head on now will, quite lterally, only store up greater problems.

The products referenced in this site are provided by parties other than BTC. BTC makes no representations regarding either the products or any information about the products. Any questions, complaints, or claims regarding the products must be directed to the appropriate manufacturer or vendor. Click here for usage terms and conditions.

©2006 BTC. All rights reserved.
No part of this site may be reproduced without written permission of the owners.
For Technical problems with this site contact the Webmaster