> > > > Ticket Details > > =================== > > Ticket ID: XKA-295690 > > Department: Support IDV > > Priority: Normal > > Status: Closed > > > One small query, every time create a new display control. unpackfloats > with copy=true gets called up ignoring the value of > arguement copy coming in in the following constructor as arguement: > public CachedFlatField(CachedFlatField that, boolean copy, > FunctionType type, Set domainSet, > CoordinateSystem rangeCoordSys, > CoordinateSystem rangeCoordSysArray, > Set rangeSets, Unit units); > > The partial stack trace is as follows: > visad.data.CachedFlatField.<init>(CachedFlatField.java:171) > visad.data.CachedFlatField.cloneMe(CachedFlatField.java:214) > ucar.visad.Util.clone(Util.java:1228) > ucar.unidata.data.grid.GridUtil.setParamType(GridUtil.java:3041) > ucar.unidata.data.grid.GridDataInstance.setParamType(GridDataInstance.java:404) > ucar.unidata.data.grid.GridDataInstance.getGrid(GridDataInstance.java:587) > ucar.unidata.idv.control.DisplayControlImpl.getGrid(DisplayControlImpl.java:3448) > ucar.unidata.idv.control.DisplayControlImpl.getGrid(DisplayControlImpl.java:3431) > ucar.unidata.idv.control.PlanViewControl.loadDataAtLevel(PlanViewControl.java:1174) > ucar.unidata.idv.control.PlanViewControl.loadDataAtLevel(PlanViewControl.java:1142) > ucar.unidata.idv.control.PlanViewControl.setData(PlanViewControl.java:699) > ucar.unidata.idv.control.ImagePlanViewControl.setData(ImagePlanViewControl.java:114 > > I mean to say a fresh new copy of flatfield gets created or atleast a > fresh new copy of its range samples gets created > in addition to the one created which is returned by getDataInner() > function of the DataSource classes. > As a result of this the memory spike goes for a high jump. > A sample test run says that if I can display an 11.2 K x 11.2 K image in > 1.3 GB of RAM, still it needs 1.7 GB of RAM only due to this. > Although after GC, it stands at 800 MB RAM. > It may a design decision but just out of curosity I wanted to know what > is the reason behind this. > thanks for listening to me patiently. In many situations, the IDV need to create a copy when doing the calculation and set this parm to true. Do you really need a fresh copy? I am not surprise for the memory spike if as you are dealing with large dataset. Yuan > I am marking copy of this mail to Don Sir too as I discussed it with him > too but he was a bit busy. May be he can also help us in > understanding this. > > regards > Ghansham > > > > Ticket Details =================== Ticket ID: XKA-295690 Department: Support IDV Priority: Normal Status: Closed
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.