There are two simple methods used in compression. Those are lossless and lossy. Lossy compression shrinks the size of a file by deleting bits of the file that won't dramatically change the quality of information stored, at least on the human level. Lossy compression is most commonly used in photos or MP3 files. Data owners may choose to trade in the integrity of their original data and information for more storage, but infrastructures rarely carry the liberty of trading quality for disk space, as quite often the data given to them is entrusted with the stipulation that it is kept safe. Because of this, lossless compression may be in demand from a customer.

The oldest storage compression is called traditional compression. This is used best on text files and other similar sources. The compression manager analyses a small segment of data, searching for patterns that can be reduced. The compression engine can be implemented in hardware or software.

When making a choice, almost all cases favor a hardware compression rather than software compression. Having both, however is strongly advised against. It will have little to no effect on the compression, and will indeed slow everything down.

Data deduplication is like traditional compression, with the exception that it functions on larger datasets, erasing duplicate chunks of data under it's control. The deduplicated data is often then compressed using a traditional compression technique. The space required when storing deduplicated data is dependant on the redundancy of the data. Data deduplication was available only at a file level until recently. With more processing abilities, smaller pieces are being deduplicated, down to a byte level.

Data deduplication is gaining speed in the modern days of computing. The swells of data and information pooling around in every datacenter is making deduplication the most important software development to hit users of storage technology in a decade. As more companies scramble to take a corner of the market and capitalize on the deduplicating frenzy, more host software, intelligent fabrics, and disk arrays designed to implement new and sophisticated data deduplication processes and compression mechanisms will be sure to emerge.