DB2 v Daytona

Hi Everyone,

I was helping my wife with her MLIS database class last night and happened to read about AT&T's Daytona DBMS. Has anyone else heard of it? Sounds pretty impressive --

"executes complex queries against billions of records within minutes" (http://tiny.cc/PpsGr)

"Principal uses of Daytona in AT&T: Hawkeye, the database of record for all of AT&T's call detail, consisting of over 312 terabytes of information and 1.88 trillion records." (http://www.research.att.com/viewProject.cfm?prjID=69)

"Update: as of June 2007, Daytona is managing over 2.8 trillion records in this same data warehouse, with over 938 billion records in the largest table." (http://www.research.att.com/~daytona/)

One of the interesting things I read was

"Another implication of this architecture is that Daytona has no database server processes! In fact, it has no daemon processes of any kind. Every query executable is on its own to run and produce its answers. Most other DBMS have invested quite a bit of effort into creating database server processes, which provide many services including proprietary file systems, scheduling, caching, locking, parallelization, security, networking, and of course, query optimization and execution. Notice that with the exception of the last two, all of these services are provided by modern day operating systems to one degree/flavor or another. Instead of implementing another operating system, Daytona cuts out the middleman and in effect, uses the UNIX operating system itself as Daytona’s server process. (Interestingly, OracleÔ has recently announced a product direction based on the opposite approach, which is, essentially, to get rid of the platform’s operating system itself!)" (http://www.sigmod.org/sigmod/sigmod99/eproceedings/papers/greer.pdf)


I don't know enough about Unix/Linux or Sun's OS to know how this compares to DB2. Anyone?

*Peter Dow* /
Dow Software Services, Inc.
909 793-9050
pdow@xxxxxxxxxxxxxxx <mailto:pdow@xxxxxxxxxxxxxxx> /