Emtec »

Accueil Adviser Articles Emtec Adviser - Putting the Squeeze on Data Storage Growth

Emtec Adviser - Putting the Squeeze on Data Storage Growth

Real-time compression reduces primary storage data volumes without performance degradation.
 

Data growth was cited as one of the top three data center challenges by 47 percent of respondents in a recent Gartner survey of representatives from 1,004 large enterprises. It’s no surprise, then, that organizations are looking for ways to reduce the hardware, software, administration and maintenance costs associated with storage. Sixty-two percent of survey respondents reported that they will be investing in data archiving or retirement by the end of 2011. Other high-ranking IT projects that will be employed to respond to data growth included storage consolidation, storage management, and data reduction techniques.

In the short term, data growth can be managed by freeing trapped, underutilized storage through consolidation projects. Long-term, however, organizations need to find ways to reduce the amount of data they store — particularly in high-cost Tier 1 storage.

Data reduction technologies have become commonplace in backup and archival solutions, but have been slower to reach the primary storage tier. However, new real-time compression technologies can deliver dramatic reductions in primary storage utilization without the drawbacks that have plagued older techniques.

Real-time compression can enable organizations to reclaim existing storage capacity, achieve faster throughput, support high availability configurations and improve forecast storage resource requirements. It is redefining the economics of storage, enabling customers to significantly reduce their storage footprint while improving the performance of their storage infrastructure.

Reducing the Load

Data de-duplication has become popular as a means of reducing data storage requirements in backup and archival operations. Also known as global compression, commonality factoring and referential integrity, data de-duplication eliminates redundant copies of data to reduce storage costs and shrink backup and recovery times. By focusing on backup and archival, however, data de-duplication technologies fail to apply data reduction techniques where data originates — in Tier 1 or primary storage.

Primary storage is for active data that is frequently accessed by both end-users and key applications. It is the most expensive storage tier, requiring high performance, low latency and high availability. It is also the source of the vast amounts of data that organizations generate and use to run their businesses.

The overwhelming majority of data begins life in primary storage that is used to host mission-critical software such as databases, email and transaction processing applications. But it’s not just generated once — it is replicated, distributed, warehoused, backed up and ultimately archived.to other storage tiers. If an organization can reduce its primary storage footprint, it not only saves capacity in Tier 1 but creates a waterfall of capacity throughout the storage infrastructure and cost savings across the data lifecycle.

Unfortunately, data de-duplication is not suited to primary storage. While de-duplication may result in up to 20:1 reductions in backup and archival data, it yields just 2:1 reductions in primary storage at best. Worse, data de-duplication simply requires too much performance overhead to be used in Tier 1.

Overcoming Challenges

Data compression has not been widely adopted in primary storage either because performance concerns were more critical than capacity savings. Simply put, Tier 1 data reduction must not impact performance. 

Real-time, random access compression/decompression technology, however, delivers the compression ratios and performance needed for primary storage data reduction. So-called lossless data compression maintains reliable and consistent performance and data integrity through a compressed file format that preserves all the information needed to access or re-create the original data. Unlike data de-duplication, in which files are replaced with pointers that may point to the wrong reference, lossless compression ensures that data integrity is not compromised.

Real-time compression is different from traditional compression technologies in that it can drastically reduce the amount of data that must be stored — from 50 percent to 90 percent. It enables IT managers to reduce the Tier 1 storage footprint in the data center and, as a result, reduce associated power and cooling costs and administrative overhead.

Real-time compression does for primary storage what de-duplication has done for backup. It transparently compresses primary storage without changes in performance, storage, applications, networks or processes. It overcomes the challenges that have inhibited the adoption of data reduction technologies in the primary storage tier, delivering dramatic savings without changing downstream storage processes such as replication, backup and archiving.

Data growth and the storage costs associated with that growth remain major challenges for many organizations. Real-time compression attacks the problem at the source — primary storage — and in so doing delivers real cost savings and performance gains throughout the storage infrastructure.