2015-09-28 12:01 GMT+02:00 Shulgin, Oleksandr <oleksandr.shul...@zalando.de> :
> On Sun, Sep 27, 2015 at 8:05 AM, Pavel Stehule <pavel.steh...@gmail.com> > wrote: > > the preparing of content before execution is interesting idea, that can be >> used more. The almost queries and plans are not too big, so when the size >> of content is not too big - less than 1MB, then can be used one DSM for all >> backends. >> > > >> When size of content is bigger than limit, then DSM will be allocated >> specially for this content. The pointer to DSM and offset can be stored in >> requested process slot. The reading and writing to requested slot should be >> protected by spinlock, but it should block only two related processes for >> short time (copy memory). >> > > Sorry, I don't think this will fly. > > The whole idea is that a backend publishes the plan of a query just before > running it and it doesn't care which other backend(s) might be reading it, > how many times and in which order. The only required locking (implicit) is > contained in the code for dsm_attach/detach(). > I didn't propose too different solution. There is only one difference - sharing DSM for smaller data. It is similar to using usual shared memory. Regards Pavel > > >> Other possibility is showing the size of content in requested process >> slot. Then the requester can preallocate enough size of shared memory. But >> this doesn't solve a issues related to waiting requester for content. So >> first variant is pretty simple, and should be preferred. The disadvantage >> is clear - it can enforce maybe significant slowdown of fast queries. >> > > Both of these approaches have just too many synchronization problems, IMO. > > -- > Alex > >
