indexes

cost model - There are B data pages, R records per page. 
	     Read/write for a pages is D, Avg process time for record is C.

Organizations
Heap - good storage efficiency, fast scan, insertion, deletion, slow search

Sorted good storage, slow inset & delete, FAST search, Best range select.

Hash - not as good space, fast insert & delete, super fast equality select
	no range support, slow full file scans, more pages per file.

*-An index is clustered  if the ordering of the data entries in the index 
is the same or close to the ordering of the data records.
*-Unclusteered index isn't clustered,, but we can have more then one per 
  data file.


*-An index is sparse if if it contains one entry for each page or records 
in the data file.
*-An index is dense if it contains at least one data entry for every 
search key value that appears in a record in the indexed file.

Primary VS Secondary indexes. An index on a set of fields that contains 
the primary key is a primary index. A secondary index is a not primary 
index.

Indexing strategy for Apps-in notebook.


Transcation-a transaction is a sql statement, or a set of statements that 
bring the database from one logical state to another. A transaction is 
only complete afeter it sends a commit call to the dbms kernel, otherwise 
it is 'rolledback' 

ACID






Concurrency




Schedules





Serializabilty






conflicts



2pl protocal-two phase locking protocal-






*P&D dbs

Def & Charastics of P & D




SMP, shared nothing




sclae-up




speed-up




client-server



middleware



collaborating server arch




replication - sync and async





2pc protocal