Search Results

Search found 3 results on 1 pages for 'user293594'.

Page 1/1 | 1 

• MySQL - how long to create an index?

  - by user293594

  Can anyone tell me how adding a key scales in MySQL? I have 500,000,000 rows in a database, trans, with columns i (INT UNSIGNED), j (INT UNSIGNED), nu (DOUBLE), A (DOUBLE). I try to index a column, e.g. ALTER TABLE trans ADD KEY idx_A (A); and I wait. For a table of 14,000,000 rows it took about 2 minutes to execute on my MacBook Pro, but for the whole half a billion, it's taking 15hrs and counting. Am I doing something wrong, or am I just being naive about how indexing a database scales with the number of rows?

  Read the article

• MySQL - optimising selection across two linked tables

  - by user293594

  I have two MySQL tables, states and trans: states (200,000 entries) looks like: id (INT) - also the primary key energy (DOUBLE) [other stuff] trans (14,000,000 entries) looks like: i (INT) - a foreign key referencing states.id j (INT) - a foreign key referencing states.id A (DOUBLE) I'd like to search for all entries in trans with trans.A 30. (say), and then return the energy entries from the (unique) states referenced by each matching entry. So I do it with two intermediate tables: CREATE TABLE ij SELECT i,j FROM trans WHERE A30.; CREATE TABLE temp SELECT DISTINCT i FROM ij UNION SELECT DISTINCT j FROM ij; SELECT energy from states,temp WHERE id=temp.i; This seems to work, but is there any way to do it without the intermediate tables? When I tried to create the temp table with a single command straight from trans: CREATE TABLE temp SELECT DISTINCT i FROM trans WHERE A30. UNION SELECT DISTINCT j FROM trans WHERE A30.; it took a longer (presumably because it had to search the large trans table twice. I'm new to MySQL and I can't seem to find an equivalent problem and answer out there on the interwebs. Many thanks, Christian

  Read the article

• Using Partitions for a large MySQL table

  - by user293594

  An update on my attempts to implement a 505,000,000-row table on MySQL on my MacBook Pro: Following the advice given, I have partitioned my table, tr: i UNSIGNED INT NOT NULL, j UNSIGNED INT NOT NULL, A FLOAT(12,8) NOT NULL, nu BIGINT NOT NULL, KEY (nu), key (A) with a range on nu. nu ought to be a real number, but because I only have 6-d.p. accuracy and the maximum value of nu is 30000. I multiplied it by 10^8 made it a BIGINT - I gather one can't use FLOAT or DOUBLE values to PARTITION a MySQL table. Anyway, I have 15 partitions (p0: nu<25,000,000,000, p1: nu<50,000,000,000, etc.). I was thinking that this should speed up a typical to SELECT: SELECT * FROM tr WHERE nu>95000000000 AND nu<100000000000 AND A.>1. to something of the order of the same query on a table consisting of only the data in the relevant partition (<30 secs). But it's taking 30mins+ to return rows for queries within a partition and double that if the query is for rows spanning two (contiguous) partitions. I realise I could just have 15 different tables, and query them separately, but is there a way to do this 'automatically' with partitions? Has anyone got any suggestions?

  Read the article

1