morgo

31 ms·

271.
by morgo 16y ago
Do you really think Monty had anything to do with coding the website anyway? The guys that write the actual database have nothing to with the web team. They didn't in MySQL days, they don't now.
272.
Video: The InnoDB Storage Engine for MySQL(http)
1 points by morgo 16y ago|1 comments
273.
by morgo 16y ago
Each modification is going to be atomic, but you don't have transactions. It still has durability against partially written writes via InnoDB's double write buffer - so it's more ACID than most NoSQL solutions.
274.
by morgo 16y ago
Right: It's very workload sensitive how this is going to help. If your data set fits in memory, it's probably very good. If you're waiting on IO, the cost of SQL parsing becomes trivial in comparison. What is actually interesting, is that
275.
by morgo 16y ago
The poster corrected himself above - it's not the case.
276.
by morgo 16y ago
It's a composite index. A covering index would be this: ALTER TABLE staff ADD INDEX idx (last_name, first_name); SELECT first_name FROM staff WHERE last_name = 'n'; A covering index is something very special - all results are returned from
277.
by morgo 16y ago
You can get a profile (strace like) in the Percona-enhanced releases: http://www.percona.com/docs/wiki/percona-server:features:pro... There is likely some overhead for this feature, as it needs to make more system calls (getrusage).
278.
by morgo 16y ago
If we want to 100% correct, indexes don't just hurt writes, and they don't hurt all writes either. Two examples (there are more): - If I do an UPDATE query, and I don't modify indexed columns, the index doesn't need to be modified. - If I
279.
by morgo 16y ago
The example is the wrong case. Unless you are using MyISAM tables (hint: 90%+ of the time you shouldn't), there's rarely a situation to LOCK TABLES as described. Using transactions would make more sense: try { mysql_query("START TR
280.
by morgo 16y ago
What irritates me about this thread is the sense of entitlement from people who file bugs. If helly wants to say "there was a discussion about this on the internals list, and the decision was no" he doesn't have to justify anything more. If
281.
by morgo 16y ago
I think it is very workload dependent how big of a power consumer the hard-drive is. When I only do word processing or browsing simple web pages, I never have to read data (everything is cached). Laptop-mode also delays writing data back.
282.
by morgo 16y ago
I don't see how your comparison is remotely fair to Twitter. Youtube is just a website, if they wanted to change how they allocate video numbers, nobody except for rogue bots will notice. Twitter is an API. Even if they have the knowledge
283.
by morgo 16y ago
-1. With hindsight it's easy to make a comment like yours, but many clients rely on the status_id to be monotonic. You would be changing behavior in a way that breaks apps. Twitter doesn't want to do that.
284.
by morgo 16y ago
Almost. RAND() writes the rand seed as meta data to the binary log. It's safe (as is NOW() - like you pointed out). UUID() is not safe though. But examples like this are so special case - it doesn't show the real danger. A DELETE with a
285.
by morgo 16y ago
"No benefit for HDD bound workloads" This means no benefit over regular MySQL releases, not to say that MySQL doesn't have some nice features. InnoDB can read/write in multiple threads - which is useful on RAID controllers which require con
286.
by morgo 16y ago
Yes, InnoDB was standalone. But it had it's own SQL parser and things like Stored Procedures as well ;) It was rudimentary, but I don't know if KVS is the best description. How MySQL and InnoDB talk is very much a K<->V / row base
287.
by morgo 16y ago
The concept of variance reduction is not even new with computing - In manufacturing they have six sigma. I believe the famous GE CEO's quote was "Our Customers Feel the Variance, Not the Mean" (Google it). This is talked about here: http
288.
by morgo 16y ago
I have 8GB RAM + an Intel SSD in my MacBookPro. There are very few physical reads - most of the data I require fits in filesystem cache. When there are physical reads, the response time is awesomely fast. An IO is less than 1ms. For the
289.
by morgo 16y ago
For 5.1 OurDelta are MariaDB builds. So there are only two to consider. See this thread on Stackoverflow, and my reply: http://stackoverflow.com/questions/2450534/which-mysql-fork-...
290.
by morgo 16y ago
If they understand how to use memcached, but they can't do: SELECT col1, col2 FROM mytable; .. then something is seriously wrong with the world.
291.
by morgo 17y ago
Except for CPU scalability, most of the new features require configuration to be turned on. Some of the features are just that they allow something to be configurable and nothing else. For example in the top 10 article linked to, being a
292.
by morgo 17y ago
To answer your question: Yes - I've used it. I wrote the first article you linked to though, so I guess I had to :p