Pages, Bytes, and Other Units of Measure

To see how memory is being used, run the Pages Memory Monitor. On MPE V disc space was managed in Sectors of 256 bytes and MPE/iX often refers to sectors as well. Just remember that a software page is 16 sectors.

Sector

Extents are a feature of MPE/iX left over from MPE V. They can almost, but not completely, be ignored. On MPE V, all extents must be the same size (except the last) and the maximum per file is 32. This can make it hard to find big enough chunks of free disc space for a large data file. On MPE/iX a file can have unlimited extents and they don't have to be the same size. Many MPE/iX files, such as Obj files, have a Record size of 256 bytes, a Record Limit of 4,096,000, and unlimited extents.

Stan Sieler points out that extents are not actually unlimited:

While there is no hard-coded limit, it turns out that the number of extents for a single file is limited at least by the maximum size of a Label Table (because all of the extents are recorded in the Label Table). A Lable Table has no hard-coded entry limit, but is limited to a maximum of 20 extents, all on the same disc drive. Anyway, I conjecture that an approximate limit on the number of extents (assuming a big disk drive, and no other files in the Label Table!) is about 15,728,100 extents. :)

Transient disc space is space managed by MPE for user data stacks and heaps, and other temporary data structures. It can consume up to 1 gigabyte per process, especially if you link your file with the default NM stack size of 96 megabytes (this allocates at least one 2048-sector piece of transient space, even for a CM program that never uses the NM stack). See SYSGEN for maximum NMHEAP and MAXNMSTACK. If the sum of these is greater than 1 GB, then 1 GB is the limit, since this space is accessed through Space Register 5. Transient space is similar to Swap Space on HP-UX.

HP-UX measures space in 1024-byte blocks in bdf, but 512-byte blocks in df. The df command is a standard UNIX report of free disc space. bdf is a nicer Berkeley form of df supplied with HP-UX; it shows free and used disc space on each file system (it does not include Swap Space and other overhead disc space). Here is some sample bdf output:

Text version.

Filesystem	kbytes	used	avail	capacity	mounted on
/dev/vg00/lvol1	104097	66657	27030	71%	/
/dev/vg00/lvol5	79449	21	71483	0%	/tmp
/dev/vg00/lvol4	99385	64732	24714	72%	/users
/dev/vg00/lvol3	496865	418420	28758	94%	/usr
dumbo.robelle.com:/sys/apps	627120	455712	171408	73%	/pcserver/sys/apps
/dev/dsk/c3d0s2	546660	546660	0	100%	/cdrom1
/dev/dsk/c4d0s2	321402	321402	0	100%	/cdrom2

The /vg00 file systems are logical subdivisions of the hard drive allocated to different subdirectories under the root directory. The idea is similar to account and group limits in MPE. The dumbo file system is a LAN PC server, mounted as a Network File System on UNIX. And the last two are CD-ROM drives with the HP-UX and MPE manuals mounted. The capacity column is sort of equivalent to MPE/iX's max Perm percent, in that 100 percent doesn't mean that the disc is full. HP-UX generally makes the disc look 100 percent full when it is actually about 90 percent full. The remaining 10 percent is given only to superuser.

bdf reports kilobytes free per HP-UX file system. See /etc/checklist/ for all file systems known to your HP-UX system and /etc/mnttab/ for all file systems which are mounted. Disc space is managed as "device-files"; see /dev/dsk/* for all disc devices. When you run out of disc space, the error you see is "No space left on device." To find out who is using the disc space on HP-UX, try du -s /users/*. This shows how many 512-byte blocks are used by each subdirectory under /users.

Fragmentation

To observe and correct MPE fragmentation on MPE, use the De-Frag/X product from Lund Performance Software or use the Contigvol command of MPE/iX 5.0's Volutil program (it creates contiguous free disc space on a volume; Contigvol appears to work about as well as VINIT CONDense did -- that is, it's stable and reliable, but requires multiple passes to get the best results).

To learn all about fragmentation, see Stan Sieler's paper "Analysis and Correction of Fragmentation Problems." Here's an example of the De-Frag/X display of disc drive fragmentation. (You can also zoom in on a single chunk of the display.)

:run defragx.pub.lps

De-Frag/X  A.03c (run on  3/02/95 18:49:03.4)
Copyright (c) 1993 Lund Performance Solutions

MPE/iX 5.0 (or later)

# Boots since INSTALL          :       236
Total # of mounted disks       :         6

Size of largest currently mounted disk = 163,735 pages (639 MB).

Note: system appears to have been started by a SOFT-boot (^B TC).

De-Frag/X> map 11
------------------------------------------------------------------------
Ldev   11:   (Each chunk represents 168 pages, or  .7 MB)
   [PPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPP]  0
   [PPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPP]  1
   [PPPPPPPPPPPPPPPPPPPPPPPPPppPPpppppPPPPPPPPPPPPPPPPPPPPPPPPPPPPPP]  2
   [PPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPpPP]  3
   [PPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPP]  4
   [PPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPP]  5
   [PPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPP]  6
   [PPPppPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPP]  7
   [PPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPPP]  8
   [PPPPPPPPPPPPPPPPPPPPp*p*************ppPp************************]  9
   [*********************************pPPPPPPPPPPPPPPPPPPPPPPPPPPPPPp] 10
   [****************************************************************] 11
   [****************************************************************] 12
Col[0....+....10...+....20...+....30...+....40...+....50...+....60..] Row
Available Permanent Disk:  134 MB (disk size:  544 MB)
MAP_characters:
   Free       *   Permanent  P   Transient  T   unmovable  X
   part Perm  p   part Trans t   same_ldev  x

Disc "Megabytes" Versus RAM Megabytes

IEC Standards: Binary versus Decimal

Name   Symbol  Origin       Factor
kibi   Ki      kilobinary   2 on power of 10
mebi   Mi      megabinary   2 on power of 20
gibi   Gi      gigabinary   2 on power of 30
tebi   Ti      terabinary   2 on power of 40
pebi   Pi      petabinary   2 on power of 50
exbi   Ei      exabinary    2 on power of 60

National Institute of Standards and Technology:

"Once upon a time, computer professionals noticed that 2^10 was very nearly equal to 1000 and started using the SI prefix "kilo" to mean 1024. That worked well enough for a decade or two because everybody who talked kilobytes knew that the term implied 1024 bytes. But, almost overnight a much more numerous "everybody" bought computers, and the trade computer professionals needed to talk to physicists and engineers and even to ordinary people, most of whom know that a kilometer is 1000 meters and a kilogram is 1000 grams."

"Then data storage for gigabytes, and even terabytes, became practical, and the storage devices were not constructed on binary trees, which meant that, for many practical purposes, binary arithmetic was less convenient than decimal arithmetic. The result is that today "everybody" does not "know" what a megabyte is. When discussing computer memory, most manufacturers use megabyte to mean 2^20 = 1 048 576 bytes, but the manufacturers of computer storage devices usually use the term to mean 1 000 000 bytes. Some designers of local area networks have used megabit per second to mean 1 048 576 bit/s, but all telecommunications engineers use it to mean 106 bit/s. And if two definitions of the megabyte are not enough, a third megabyte of 1 024 000 bytes is the megabyte used to format the familiar 90 mm (3 1/2 inch), "1.44 MB" diskette. The confusion is real, as is the potential for incompatibility in standards and in implemented systems."

Prof Russ Rowlett UNC:

"5 gigabytes (GB) should mean exactly 5 000 000 000 bytes, and 5 gibibytes (GiB) should mean exactly 5 368 709 120 bytes. The fate of this innovation is uncertain. So far, very few people are using the IEC binary prefixes. Searches for them on the Internet turn up, for the most part, complaints by people who don't want to use them."

freedictionary.com:

"As of 2003, this naming convention has not yet gained widespread use. The IEC did not give names for the prefixes beyond exa-, but if they had given them names, they would probably be zebi- and yobi- as well as nobi- and dogbi-."