EMC has announced block data compression for the EMC CLARiiON® and EMC Celerra® storage systems, with a claimed potential to reduce the online data by 50%. This function is very similar to NetApp’s A-SIS. A hash signature for each block is created and stored with the array metadata. An asynchronous (batch) job is run to detect blocks with the same has signature, and free up the data blocks to be reused. When the compressed data is read, there is some overhead as the data is retrieved from another location.
The ASIS virtualization architecture (based on WAFL) allows the freed data to be easily reassigned. It will be important to read the fine print in the EMC documentation when it becomes available to understand if there any limitations on compression in terms of performance, use cases, etc.
Bottom line: Compression of online data has arrived and will become a standard feature on all arrays. It will be particularly useful for flash drives to offset the higher costs of these devices.
Action Item: NetApp and EMC have enabled compression of online data to become a standard storage feature. Most arrays vendors will follow. CIOs should ensure there is a strategy to adopt on-line data compression and include EMC, NetApp and compression players like IBM Real-time Compression on their shortlist to understand the differences in technologies and their appropriateness for specific use cases.