Featured post
ado.net - Accessing SQL Server in parallel -
i'm trying use task-parallel-library offload expensive ado.net database access ui thread (formerly program i'm re-writing freeze, updating vb6 text box progress, until data in database loaded). have complex dependency structure (26 individual tasks), , i'm trying figure out how of worth parallelizing.
i'd know whether or not io access can parallelized @ performance bonuses. if not i'll sequentially load data , update ui whenever enough information loaded perform task, it'd nice boost loading maybe 2 things @ time instead of 1 (even if don't double speedup).
it's possible parallelizing increase performance, not guaranteed. depends on bottleneck is.
for example, if request expensive because loads lots of data, consumes of clients network bandwith. parallelizing in case wouldn't much, if @ all.
if, on other hand, bottleneck sql processing and sql request leaves sql server spare capacity in own bottleneck, can profit sql servers (very good) parallelizing capabilities.
it possible parallelizing slows down. if example sql server has not ram , access single disk, forcing multiple queries in parallel may lead more seek activity on harddisk, can dramatically slow down overall read rate.
so, is, answer isn't simple yes or no, "it depends".
- Get link
- X
- Other Apps
Comments
Post a Comment