Comments on: Mike Stonebraker explains column-store data compression

By: Mark

Mark — Sun, 25 Mar 2007 11:06:23 +0000

Teradata is a row store, yet it compresses by column domain (dictionary compression) and does not do block decompression on access. Access to the field is more efficient with compression because the CPU savings for IO is greater than the indirection to the value dictionary. There is a slight CPU overhead on the update side to search the value dictionary for the index to put in the row header, but this is minor in comparison to the overall win in space savings and scan rate.

By: DBMS2 — DataBase Management System Services»Blog Archive » Will database compression change the hardware game?

Sat, 24 Mar 2007 08:43:36 +0000

[...] I’ve recently made a lot of posts about database compression. 3X or more compression is rapidly becoming standard; 5X+ is coming soon as processor power increases; 10X or more is not unrealistic. True, this applies mainly to data warehouses, but that’s where the big database growth is happening. And new kinds of data — geospatial, telemetry, document, video, whatever — are highly compressible as well. [...]

By: DBMS2 — DataBase Management System Services»Blog Archive » Mike Stonebraker on data compression — comments

Sat, 24 Mar 2007 05:19:37 +0000

[...] In my opinion, the key part of Mike Stonebraker’s fascinating note on data compression was (emphasis mine): [...]