MemSQL.. Explored yet?
-
was reading an article about memSQL ... seems to be good and new thing.. but wonder does it make any difference with other documented databases?
Didn't know the specific brand; TimesTen (from Oracle) works as advertised, I assume this one does the same. Not very useful for documents - it is useful for high traffic data.
Bastard Programmer from Hell :suss: If you can't read my code, try converting it here[^][](X-Clacks-Overhead: GNU Terry Pratchett)
-
was reading an article about memSQL ... seems to be good and new thing.. but wonder does it make any difference with other documented databases?
memSQL is still a RDBM, but uses memory as storage, while document-base is noSQL...so not much to compare... If you are interesting in hi-performance (real-time) data processing, you may check SQL's in-memory-tables...
Skipper: We'll fix it. Alex: Fix it? How you gonna fix this? Skipper: Grit, spit and a whole lotta duct tape.
-
memSQL is still a RDBM, but uses memory as storage, while document-base is noSQL...so not much to compare... If you are interesting in hi-performance (real-time) data processing, you may check SQL's in-memory-tables...
Skipper: We'll fix it. Alex: Fix it? How you gonna fix this? Skipper: Grit, spit and a whole lotta duct tape.
-
was reading an article about memSQL ... seems to be good and new thing.. but wonder does it make any difference with other documented databases?
Schatak wrote:
seems to be good and new thing
Pretty sure when new things stop showing up in computing then people won't be doing computing any more. So new doesn't mean much. The real question does it serve a real need significantly better than some older technology. Must be significant otherwise the learning time and the problems due to learning will not be an effective trade off. And the vast majority of the time for any really "new" idiom the answer is no. So based on that one shouldn't look for new technologies but rather look for solutions to real problems. For example a real problem might be that an existing file based database using expertly created caching and expertly created database design fails to meet real performance needs. This would be versus situations like the following - My 'database' is slow, when no actual profiling has been done at all. (I don't even know what profiling is much less creating a simulation of actual production traffic.) - My database is slow but I don't have any idea how databases work. - I don't like SQL and consequently all my SQL ends up being written like it didn't exist. - I don't even know what caching strategies are - I read about this cool tool and I want to base the entire future of a company for the next 10 years on it just so I can play with it. - As a developer I need to restore the database 20 times a day and it just takes too long so the database should be replaced even though there is no problem with performance in production.
