mirror of
https://github.com/kristoferssolo/Databases-II-Cheatsheet.git
synced 2025-10-21 18:20:35 +00:00
Compare commits
11 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 29d1784813 | |||
| 1cd6f44af8 | |||
| 97934dd5ac | |||
| f3345f7cc1 | |||
|
|
f6128287ad | ||
| 1bf92654da | |||
| 3f4261e918 | |||
| 928bd8554f | |||
|
|
a9bb553cf3 | ||
|
|
226bf240d6 | ||
|
|
5f19876f7a |
BIN
export/DB2Cheatsheet.pdf
Normal file
BIN
export/DB2Cheatsheet.pdf
Normal file
Binary file not shown.
22
main.typ
22
main.typ
@ -87,9 +87,10 @@ Simplified version (to get the idea)
|
||||
for each tuple tr in r: (for each tuple ts in s: test pair (tr, ts))
|
||||
```
|
||||
|
||||
// TODO: Add seek information
|
||||
Block transfer cost: $n_r dot b_s + b_r$ block transfers would be required,
|
||||
where $b_r$ -- blocks in relation $r$, same for $s$.
|
||||
where $b_r$ -- blocks in relation $r$, same for $s$. Each scan of the inner
|
||||
relation requires one seek, and the scan of the outer relation requires one seek
|
||||
per block, leading to a total of $2 dot b_r$ seeks.
|
||||
|
||||
== Block-nested join
|
||||
|
||||
@ -275,7 +276,8 @@ The concept of conflict equivalence leads to the concept of conflict
|
||||
serializability. We say that a schedule $S$ is *conflict serializable* if it is
|
||||
conflict equivalent to a serial schedule.
|
||||
|
||||
=== Serializability graph
|
||||
// TODO: rename to precedence
|
||||
=== Precedence graph
|
||||
|
||||
Simple and efficient method for determining the conflict seriazability of a
|
||||
schedule. Consider a schedule $S$. We construct a directed graph, called a
|
||||
@ -310,6 +312,15 @@ locking protocol. We say that a locking protocol ensures conflict
|
||||
serializability if and only if all legal schedules are *conflict serializable*;
|
||||
in other words, for all legal schedules the associated $->$ relation is acyclic.
|
||||
|
||||
*Recoverable* schedule is one where, for each pair of transactions $T_i$ and
|
||||
$T_j$ such that $T_j$ reads a data item previously written by $T_i$, the commit
|
||||
operation of $T_i$ appears before the commit operation of $T_j$.
|
||||
|
||||
*Cascadeless* schedule is one where, for each pair of transactions $T_i$ and
|
||||
$T_j$ such that $T_j$ reads a data item previously written by $T_i$, the commit
|
||||
operation of $T_i$ appears before the read operation of $T_j$. Every cascadeless
|
||||
schedule is also recoverable.
|
||||
|
||||
=== Lock-based
|
||||
|
||||
*Shared Lock* -- If a transaction $T_i$ has obtained a shared-mode lock (denoted
|
||||
@ -475,7 +486,8 @@ $B=5;T_"disk"=0.001;T_"seek"=0.1$
|
||||
=== Nested-Loop Join Method
|
||||
+ Nested-loop join:
|
||||
- For each pattern in $r_1$, search all patterns in $r_2$.
|
||||
+ Total Combinations: $75435 dot 11456=$
|
||||
+ Total Combinations: $75435 dot 11456$
|
||||
+ Time Calculation for Nested-Loop Join:
|
||||
- Reading and searching time for each combination: $0.001+0.1=0.101 "ms"$
|
||||
- Total time: $75435 dot 11456 dot 0.101 = 87282519.36 "ms"$
|
||||
- Total time: $75435 dot 11456 dot 0.101 = 87282519.36 "ms"$ [Authors comment:
|
||||
seems fishy]
|
||||
|
||||
Loading…
Reference in New Issue
Block a user