Samsung has announced a new line of 64G SSDs for use in consumer laptops. Nice to see innovation but I REALLY want to see more SSDs in the server space.

I JUST want to use a CHEAP 2-5G SSD for write ahead and binary logs. I just need 1k write ops per second with a SATA interface. Then I could drop them in cheap commodity servers and get a significant performance boost. My disks would still use software RAID 0 (at least until I replace this partitioning) but only store data.

Give me one of these for $50 to $100 and I’ll buy one with ever server.

The fsync’s required for writing to disks on INNODB can really hurt performance. My bet …

Well, it seems many have put up their wishlist for features they would like to see in MySQL so here is my list. These are in no particular order of importance, just as they came to mind :)

1. Online Backup - This is currently being developed and it will make a big difference to many people currently trying to administrate MySQL. We can only hope that it gets done soon and is well tested to avoid any major issues.

2. Standardised HA/Replication - Along similar lines to the consistent online backup is that of consistent high availability, load balancing and replication. No matter what the engine being used, this will allow a consistent method of implementing these features regardless.

read more

Well, it seems many have put up their wishlist for features they would like to see in MySQL so here is my list. These are in no particular order of importance, just as they came to mind :)

1. Online Backup - This is currently being developed and it will make a big difference to many people currently trying to administrate MySQL. We can only hope that it gets done soon and is well tested to avoid any major issues.

2. Standardised HA/Replication - Along similar lines to the consistent online backup is that of consistent high availability, load balancing and replication. No matter what the engine being used, this will allow a consistent method of implementing these features regardless.

read more

I am indeed amazed when I consider how weak my mind is and how prone to error --Rene Descartes

This is not a recommended approach; only explained here so that you can identify it when it presents itself and avoid it.

The following indexes are created on the star schema example shown before:

1. Concatenated index is created on the fact table: sales_transaction (transaction_detail_key, product_key, sales_org_key, calendar_key) - note the difference, compared to the nested join approach where multiple indexes on each foreign key were created
   
2. In addition some indexes may be created on the dimension attributes that are frequently queried: transaction_details (transaction_type), product (SKU), sales_organization …

One cannot have everything the way he would like it --Mark Twain

In the absence of other advanced techniques for star join optimization, nested fact join with dimensions while ensuring that the indexes are used for each dimension join delivers reasonable performance.

The following indexes are created on the star schema example shown:

1. Primary key unique index on surrogate key for each dimension. One index each is created on: calendar (calendar_key), transaction_details (transaction_detail_key), product (product_key), sales_organization (sales_org_key)
2. One index is created for each foreign key on the fact table: sales_transaction (calendar_key), sales_transaction (transaction_detail_key), sales_transaction (product_key), sales_organization (sales_org_key)
3. In addition some indexes may be …

Lookup tables contain, in general, a fixed list of data. This data doesn?t change very often in database business applications. Examples of this data could be a product list, category type, supplier list, state name, zip code, phone area code, etc. In Windows and Internet web business applications, most of these lookup tables are graphically implemented by using ComboBox, ListBox or CheckListBox read-only controls. These controls are loaded with data using two main columns, ID and Name. For example, the USA state table, the ID could be ?CA? and the Name ?California?. Some times, for standard Windows form and Internet web page we need to show data to the end-users from many of these lookup tables. A fast data loading process and defining the main column values for each lookup table is required. In this article I will show you standard lookup data loading procedure and the generic classes object to store and read-only the values of the ID and Name …

Since our goal is to find a tool for profiling while we don't want to change/modify the existing code. So we chooses the Aspect-oriented C tool as the profiling language.

The basic idea is to maintain the profiling code separately, and then use Aspect C to weave the code into the original code automatically. For instance, if we are going to profile the I/O function pread, we can write the below code according Aspect specification. Then every time when pread is called, it will be executed.

#include
long pbxt_profile_pread_access_time;
long pbxt_profile_pread_freq;

int around(): execution(int pread(...)) {
struct timeval start_time;
struct timeval end_time;
gettimeofday(&start_time,NULL);

int val = proceed();

gettimeofday(&end_time,NULL);
…

It has gotten a bit quiet around the MySQL Proxy over the last weeks, but I can assure you it was worth it.

Here in my hands I hold MySQL Proxy 0.5.0. Not just some binaries, no everything with sources. As always it is dual licensed under the GPL and commercial MySQL license. We worked hard to get everything ready for the release: the wiki, the forums, the public SVN trees ...

The proxy can do magic. Put in your ideas and say the magic words and there ... it works. :)

The proxy is fully scripted now which opens up the proxy to your ideas. The possibilities are endless. I can think of at least of the following features:

• load balancing
• fail over handling
• query analysis
• SQL macros
• query rewriting
• .... much more

The proxy started as side project of me and is now evolved into a full MySQL project maintained by the …

After years of "No Call for Papers", and "How can I speak at Fosdem ?"

Now you can suggest speakers and subjects for the 2008 Fosdem edition so let's all fill in what we want at next years edition !

But what do we want, I guess we want a Scale the Web2.0 track with talks on MySQL , Apache, Loadbalancing, System Automation etc., don't we ?