diff options
Diffstat (limited to 'storage/ndb/src/kernel/blocks/dbtux/tuxstatus.html')
| -rw-r--r-- | storage/ndb/src/kernel/blocks/dbtux/tuxstatus.html | 120 |
1 files changed, 120 insertions, 0 deletions
diff --git a/storage/ndb/src/kernel/blocks/dbtux/tuxstatus.html b/storage/ndb/src/kernel/blocks/dbtux/tuxstatus.html new file mode 100644 index 00000000000..264809cefd3 --- /dev/null +++ b/storage/ndb/src/kernel/blocks/dbtux/tuxstatus.html @@ -0,0 +1,120 @@ +<HTML> +<HEAD> +<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1"> +<TITLE>NDB Ordered Index Status</TITLE> +</HEAD> +<BODY LINK="#0000ff" VLINK="#800080" BGCOLOR="#ffffff"> +<p> +<h2>NDB Ordered Index Status</h2> +<p> +<h3>Alpha release Jan 30, 2004</h3> +<p> +<ul> + <li> + Up to 32 index attributes of any type, possibly nullable. + <li> + Index build i.e. table need not be empty. + <li> + Logging NOT done, index rebuilt at system restart. + <li> + Single range scan with lower and upper bounds. + <li> + Scan with locking: read latest, read for update. + <li> + LIMITED number of parallel scans. + <li> + Total result set NOT in index key order. + <li> + NDB ODBC optimizer to use ordered index for equality but NOT for ranges. + <li> + MySQL optimizer to use ordered index for equality and ranges. +</ul> +<p> +As an example, consider following index on integer attributes. +<p> +<tt>SQL>create index X on T (A, B, C) nologging;</tt> +<p> +Single range scan means that bounds are set on +an initial sequence of index keys, and all but last is an equality. +<br> +For example following scans are supported (the last 2 not via NDB ODBC). +<p> +<tt>SQL>select * from T where A = 1;</tt> +<br> +<tt>SQL>select * from T where A = 1 and B = 10 and C = 20;</tt> +<br> +<tt>SQL>select * from T where A < 10;</tt> +<br> +<tt>SQL>select * from T where A = 1 and 10 < B and B < 20;</tt> +<p> +Following scans are NOT supported: +<p> +<tt>SQL>select * from T where B = 1;</tt> +<br> +<tt>SQL>select * from T where A < 10 and B < 20;</tt> +<br> +<h3>Features and dates</h3> +[ Now = Jan 19 ] +<p> +<table border=1 cellpadding=1> +<tr align="left"> + <th width="40%">Feature</th> + <th width="15%">Now</th> <th width="15%">Jan 30</th> <th width="15%">Mar 01</th> <th width="15%">Never</th> +</tr> +<tr align=left> + <td>Index maintenance</td> + <td>X</td> <td>-</td> <td>-</td> <td>-</td> +</tr> +<tr align=left> + <td>Basic scan</td> + <td>X 1)</td> <td>-</td> <td>-</td> <td>-</td> +</tr> +<tr align=left> + <td>Scan bounds on nullable attributes</td> + <td>-</td> <td>X</td> <td>-</td> <td>-</td> +</tr> +<tr align=left> + <td>Scan with locking</td> + <td>-</td> <td>X</td> <td>-</td> <td>-</td> +</tr> +<tr align="left"> + <td>NDB ODBC equality bounds</td> + <td>-</td> <td>X</td> <td>-</td> <td>-</td> +</tr> +<tr align="left"> + <td>MySQL integration</td> + <td>-</td> <td>X</td> <td>-</td> <td>-</td> +</tr> +<tr align=left> + <td>Index build</td> + <td>2)</td> <td>X</td> <td>-</td> <td>-</td> +</tr> +<tr align=left> + <td>Unlimited number of scans</td> + <td>3)</td> <td>-</td> <td>X</td> <td>-</td> +</tr> +<tr align=left> + <td>Total ordering</td> + <td>-</td> <td>-</td> <td>X</td> <td>-</td> +</tr> +<tr align=left> + <td>Multiple range scan</td> + <td>-</td> <td>-</td> <td>X</td> <td>-</td> +</tr> +<tr align="left"> + <td>NDB ODBC range bounds</td> + <td>-</td> <td>-</td> <td>-</td> <td>X</td> +</tr> +<tr align=left> + <td>Logging</td> + <td>-</td> <td>-</td> <td>-</td> <td>X</td> +</tr> +</table> +<p> +1) No locking and bounds must be on non-nullable key attributes. +<br> +2) Currently table must be empty when index is created. +<br> +3) Currently limited to 11 simultaneous per fragment. +</BODY> +</HTML> |