Article View: comp.os.msdos.programmer
Article #249Re: Paging large data files
From: Ping Huang
Date: Thu, 06 Aug 1992 01:02
Date: Thu, 06 Aug 1992 01:02
32 lines
2476 bytes
2476 bytes
In article <1992Aug5.193559.19274@infonode.ingr.com> fisherm@infonode.ingr.com (Matthew Fisher) writes: > I'm dealing with a table the is larger than 640k. I need some code or > ideas on how to do effcient page swapping. Access of the table is in no > particular order. One access my be in the first 20k and the next access may > be up past 2mg. The data cannot be sorted to make the access any better. It sounds like what you really want to do is use a DOS extender and a 32-bit C compiler instead of writing a Windows program. The 32-bit C compiler part allows you to write programs which can malloc a two megabyte array, for instance, and crunch data objects larger than 64Kb without giving it a second thought. You get full access to the amount of extended memory in your PC, rather than the 1Mb you get under normal DOS programs. The DOS extender part allows your 32-bit program code to do I/O and otherwise call upon DOS services, because it translates your requests into 16-bit requests that DOS can understand. Some DOS extenders can provide virtual memory, as Windows can in 386 enhanced mode, but the DOS extender can do so with less speed overhead and takes up less of your memory than Windows does. Or you could write your program for OS/2 2.0. :) You would need a 32-bit C compiler (no DOS extender necessary, and OS/2 itself is quite capable of providing virtual memory). GNU's C/C++ compiler has been ported and works quite well for data crunching. Followups either to comp.os.msdos.programmer (for DOS extender discussions) or comp.os.os2.programmer for OS/2 2.0 discussions. -- | Ping Huang (INTERNET: pshuang@martigny.ai.mit.edu) speaking for himself. | "One Thing to name them all, One Thing to define them, | One Thing to place them in environments and bind them..."
Technical Details
Message-ID:
<PSHUANG.92Aug5210204@freeside.ai.mit.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!stanford.edu!snorkelwacker.mit.edu!ai-lab!zurich.ai.mit.edu!pshuangReferences:
<1992Aug5.193559.19274@infonode.ingr.com>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!stanford.edu!snorkelwacker.mit.edu!ai-lab!zurich.ai.mit.edu!pshuang From: pshuang@zurich.ai.mit.edu (Ping Huang) Newsgroups: comp.os.msdos.programmer,comp.sys.ibm.pc.programmer,comp.windows.ms.programmer,comp.os.ms-windows.programmer.misc Subject: Re: Paging large data files Message-ID: <PSHUANG.92Aug5210204@freeside.ai.mit.edu> Date: 6 Aug 92 01:02:04 GMT References: <1992Aug5.193559.19274@infonode.ingr.com> Sender: news@ai.mit.edu Followup-To: pshuang@athena.mit.edu Organization: M.I.T. Artificial Intelligence Lab. Lines: 33 In-reply-to: fisherm@infonode.ingr.com's message of 5 Aug 92 19:35:59 GMT