What is the maximum amount of memory any single process on
Windows can address? Is this different than the maximum
virtual memory for the system? How would this affect a
system design?
- 2 Answers
- 9370 Views
- I also Faced
- E-Mail Answers
Answer Posted / rek
What is the maximum amount of memory any single process on
Windows can address? Is this different than the maximum
virtual memory for the system? How would this affect a
system design?" NT-based OSes split memory into user and
kernel space. The amount of memory usable is dependent on
the exact flavor of the OS and whether the processor is
32-bit or 64-bit. In the 32 bit world, XP and Win2K both
support 4GB, Win2K Advanced Server supports 8GB and Win2K
Datacenter Server supports up to 32GB. XP and W2K's 4GB gets
split 50/50, with 2GB being available for a process and the
remaining 2GB being held for the kernel. (Note that this
last space isn't just for the OS -- all the hardware devices
use this space as well, so that killer graphics card you got
with gobs of video RAM uses up portions of the kernel space
as well. This is the same in 32- or 64-bit procs.) XP Pro
and Win2003 Server can use the /3GB switch in boot.ini to
make 3GB available to the user side. The 2GB limit isn't a
Windows limit, it's a 32-bit processor limit and hits Linux,
Solaris, etc. (I'm sure Macs don't have this problem because
Macs are the perfect system and have no limitations
whatsoever. Ask any Mac user.) 64 bit Windows supports up to
16 terrabytes (woo hoo, gimme one o' dem filled up!) but
still split memory 50/50 between user and kernel. 32 bit
applications are still limited to 2GB user space. 64-bit
Windows versions don't support the /3GB switch. So yes, a
process's max memory is different than the system's total
virtual memory. System design is impacted if you're trying
to create a system using very large data structures,
particularly if your software's running on a system with
other apps. Your design needs to make careful use of memory
to make sure you're not going to run out of virtual memory.
Furthermore, since virtual memory is a combination of
physical memory and swapfile space, you need to take care to
avoid paging -- or at least minimize it as much as possible.
This means you should have an understanding of the
environment your software's going to run in. At the minimum,
you should have good documentation explaining your
assumptions, estimated system requirements and projected
system impacts. The first chapter in Bentley's Programming
Pearls is a great discussion on careful use of system
memory, plus it shows the benefit of ensuring you know what
problem you're really trying to solve. The story opening the
problem relates a conversation Bently had with a friend. The
short version is that what Bently first thought was a simple
sort of a large file on disk turned out to be a horse of a
different color. The "real" problem required some creative
thought to get around
| Is This Answer Correct ? | 2 Yes | 0 No |
Post New Answer View All Answers
through which algorithm does the garbage collector works? how the garbage collector will understand that the object will going to be deleted?
WS-NUM PIC S9(05)V(02) SIGN TRAILING SEPARATE MOVE '0050000+' TO WS-NUM The value stored is 00500,00+ MOVE '0050000-' TO WS-NUM Then what is the value will be stored in WS-NUM? Am getting '-00500,00'.....>>> What should I declare to WS-NUM so that I can get correct values for both + & - signs.
why we use abstract in awt?
how many types of bytes are there???
hai i am mca 2009 fresher.please tell me which certification helps me to get an IT job faster which institute is good in hyderabad.please mail me to prasanna.1856@rediff.com
Hello Experts, What is the difference between move and move corresponding exactly? please post me asap
What for decision coverage and modified condition decision coverage are used? Wat is the difference between them?
Hai, My name is nisha.I have NIC exam.If anybody Knows NIC previous paper pattern pls send me to nishanairp@gmail.com
SAS question: I have 50 fils a1,a2,...,a50. the primary key is upc. then if i want to merge all 50 files, the code is as follows, data test; merge a1 ... a50; by upc; run; we know that writing all 50 files name is time consuming, is there any standard format of this code?
what is the difference between Windows application and Unix application?
Write a shell program. Enter number of days from keyboard. Find out the number of years, month and days it contains
How can recruiter justified that the candidate is expert in Algorithm and datastructure for Software product development ?
a characteristic of a multiprogramming is? a.simultaneous exe of pgm instr 4m 2 appli b.concurrent processing of 2 r more prgms c.multiple cpu s d.all the abov
I m Abdullah Ansari compleated MCA from Jamia Hamdard,i have appeared the test of IBM on 2 august at oxford college of engineering Bangalore.waiting for hr round.. This is the first round for IBM.02/08/08 Paper consists of 4 sections 15 questions from matrices(time very less but no negative marking). 25 questions from series completion section (this section is very easy but negative marking) 15 questions from aptitude(little bit tough time limit 15 minute negative marking) 4th section is from computer science (c,c++,operating system,digital electronics ,basic question..) result came at 3 o'clock.i was selected... In interview they asked questions like 1 they asked about final yr project.. 2 what are dynamic and static memory location? 3 linked list and array difference between them. 4 what is function ? what is difference betwen function and inline function? 5 about structure 6 about binary tree, traversal, call by value. 7 storage class and many more basic questions..
How does the TCP handle the issue of multiplexing?
