COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.REQUIRE "LET.PUB[LET,JMC]" SOURCE
C00032 ENDMK
C⊗;
.REQUIRE "LET.PUB[LET,JMC]" SOURCE
.AREA TEXT LINES 4 TO 53;
.TITLE AREA HEADING LINES 1 TO 3;
.	PAGE FRAME 53 HIGH 80 WIDE;
∂MLOT Professor G. Lieberman$Ralph Gorin and John McCarthy
$Request for more LOTS hardware$∞

.TURN OFF "$%"; TURN ON "∂";

In the academic year 1977-78, the LOTS computer facility
took over the full load of class computing (with anticipated
minor exceptions) and a substantial load of unsponsored research
and student projects.  By the end of Winter Quarter, the full
complement of authorized equipment was operating.  We believe
that the availability of interactive computing to all classes
has expanded the use of computers and increased its usefulness
for learning.  However, the computer is now overloaded,
and we have good evidence that additional computer equipment will
substantially increase the quality of service we can give.
Software improvements can also provide better service, and some
are being implemented.

We request authority to make the following acquisitions:

	1. Upgrade the processor from a 2050 to a 2060.
This involves replacing most of the electronics of the computer
and costs $50,000.

	2. Increase of memory from 512K words to 1024K words.
We have a proposal from Ampex to do this for $115,000, and a proposal
from Digital Equipment Corporation to do it for $194,000.  Digital
is also preparing another plan to do it for about $140,000.
Both vendors have offered short term leases by which we can evaluate
the effectiveness of the memory in increasing system throughput.

	3. A second tape drive at $15,000.

	4. Tax, shipping, insurance and installation charges on the above.

Total: Approximately $195,000 to $220,000, depending on our choice of
vendor.

In addition to the purchase cost, there will be an additional maintenance
cost of approximately $10,000 per year.

Adding to the main memory is the best technical solution to our
overloading problems and it can be done for Fall
Quarter.  This upgrade should increase computer throughput adequately
to serve the present level of demand from the academic users of LOTS.
In the absence of increased demands, we believe that the system with
this upgrade will deliver nearly twice as much throughput (65 to 70%
of all CPU time) under "worst-case" loads as it does presently (35%). 

Making LOTS better will tempt some people to make more extensive use of
the facility.  To restrict the growth of new uses, we have limits on
terminal time and the limited number of public terminals.  Restrictions to
limit the influx of additional classwork may be necessary; but we should
be careful neither to propose nor tolerate limitations that adversely
affect the academic program.  At some time in the future, the single LOTS
mainframe will not by itself be adequate to service all instructional
computing; when necessary, the LOTS-style of computing can be extended to
further systems.

The detailed reasons for this request follow.
.NEXT PAGE

BACKGROUND:

LOTS is operating a large DECsystem 2050 computer system to
provide interactive computing service to support instruction and
unsponsored research at Stanford University.  Of the 11,000 students
at the University nearly 3,000 use LOTS each quarter.  LOTS is used
for instruction in Computer Science, Engineering, Statistics,
Education, and the social sciences.  In addition to instruction, LOTS
is also used for unsponsored research and to some extent for
word-processing by students writing dissertations and papers.

The LOTS system hardware presently includes a DECsystem-2050
processor, 512 K words of main memory, three 200-megabyte disk drives
on two channels, one tape drive, and eighty terminal ports.

Two of the three disk drives are in use for public storage and
swapping.  The third disk is presently being used for large,
infrequently changed data files, principally system accounting data
and program sources.

THE PROBLEM:  Demand exceeds supply.

Demand for LOTS chronically super-saturates the effective supply of
services.  We must improve the ratio of effective throughput to user
demands.  The system saturates to the point where people find that it
is no longer useful.  They would quit using LOTS under those
circumstances but for external pressure from due dates.

The overload at LOTS is apparent in several effects:

.BEGIN INDENT 8,8; NOFILL
1.  Lengthy delays for response from the computer,
2.  Long queues of students waiting to use terminals,
3.  Occasional long delays for printed output,
4.  Long waits to access the tape drive, and
5.  Shortage of disk space.
.END

The first of these conditions is the most distressing.  Moreover, each of
the other symptoms of overload can be traced to the general
inefficiency of the computer system under overload conditions.

A more efficient system will give better response, get people off the
terminals quicker, shorten the queue of people waiting for terminals,
and provide better printer throughput on the printers we already
have.  Less obviously, poor system throughput is preventing us from
taking full advantage of all the disk space we already have.  (Some
further explanation is made below.)

RESTRICTIONS ON DEMAND

On the demand side, we have placed restrictions on how much console
time each user is allowed per week.  We have reason to believe that
these restrictions were not entirely effective Spring quarter.
However, a better job can be done, and it will be done this Fall.  We
are tightening up the account-making process this summer, to better
prepare for Fall quarter.

A recent episode shows how, despite our controls, we have failed to
reduce demand.  We don't want to repeat the experience  we  had  with
the Human Biology 4 course that put 200
novice users, all doing resource-consuming plots, 
onto LOTS at about the sixth week of the quarter.
However, given that LOTS is charged with
supplying all instructional computing, we have very little means to
prevent a recurrence.  When Professor Durham first approached us in
this matter, it was his intention to use the computer in early May, and
this looked reasonable to us.  However, he didn't finish his special
programs until about May 15, so the students in Human Bio 4
contributed an extra burden to the usual end-of-quarter rush.

We hope that as the faculty gains a more thorough understanding of
the capabilities and limits of the LOTS resource such incidents will
diminish in frequency and impact.  One of LOTS' functions is to
provide for innovative pedagogical use of the computer, so we must
expect that additional courses will "discover" us.  Furthermore, as
more students at Stanford become aware of the variety of services
that can be obtained at LOTS, we expect to see an increase in
the services delivered outside of required classwork.  Some other universities
have forecast their growth rate for instructional and research computing
at 35% per year.  Although the one mainframe at LOTS will not be adequate
indefinitely, the LOTS-style of computing can be extended to
further systems when necessary.

We propose to continue the console-time allocation system that we
started this spring.  There will be some changes to the system,
including the following:

.BEGIN INDENT 8,8
1. No carry-forward of unused weekly allocations to the
next week.

2. Time allocations to students in each course will be
based on the computational intensity of the particular
course, rather than on units alone.  This way, a course such as Human Biology 4 A
and B (a total of 8 units), wouldn't automatically get
an allocation of computer time twice that of CS105.

3. A tightening of the account issuing process to eliminate
questionable accounts and poaching.

4. Encouragement of some unsponsored research users to find
a little money and to avail themselves of SCIP's services.

5. More effective enforcement of the per-session time limits.

6. Establishment of a terminal reservation system, in which
a terminal can be reserved for a specific time in the
future.  This may permit people to budget their time
more effectively.

7. Possibly, extending the allocation system to directly
limiting line printer output and cpu time in addition
to terminal connect time.
.END

.NEXT PAGE
INCREASING THE SUPPLY:  An effective prescription

There are two ways to increase the supply of computer resources.
First, we can employ better software that delivers more of the
hardware system's capabilities to the users; but software techniques
alone cannot produce the drastic increase in effectiveness that LOTS
needs.  Second, we can buy additional computer hardware.
The remedy we propose includes both software and hardware.

Projects to build a terminal reservation system and
to implement the resource limitations were discussed above.  The following
projects are underway to increase the effectiveness of particular
software areas:

.begin indent 8,8
1.  The development of a new implementation of PASCAL for
use by the introductory and intermediate Computer Science
(and related Electrical Engineering) courses.  This project
is being done in response to two operational difficulties
that have been observed with the SAIL language implementation
at LOTS.  These problems with SAIL that we hope to eliminate
in the new implementation of PASCAL, are:  Incomprehensible
diagnostic error messages, which hinder students' programming
debugging, and the use of the linkage editor (LINK) which is
a large and inefficient consumer of system resources.

2. Changes to the operating system to make it allocate main
memory and other system resources more effectively.  The
LOTS staff is too small to get into this kind of tinkering
too deeply, so we are influencing DEC to make the
performance improvements that we think are necessary.
.END


Most of the increase in throughput will come from new hardware, particularly
from adding more memory.  The new hardware we seek includes:

.BEGIN INDENT 8,8;
1. Upgrade to the 2060 central processing unit,

2. Addition of 512 K words of memory, to double LOTS' main memory size,

3. Addition of one tape drive, and

4. No growth, or very small growth, in the number of
terminals attached to LOTS.  No additional terminals for
LOTS are requested at this time.
.END


The first step: a new processor

Additional memory is our most pressing need, but it is necessary
to upgrade the central processor to the 2060 level when we add memory.
The 2060 includes a faster and more
versatile CPU that can support more memory than our 2050 has. Although
we don't presently need a faster CPU, the other features of the 2060
are necessary:

.begin indent 8,8

1. LOTS needs the 2060 to keep up with DEC's future
software developments.  We must follow DEC's
operating system developments because we have insufficient manpower
to undertake the long-term development and support of such a system.

2. By extending the virtual address space, the 2060 allows us to integrate
more than two disk drives into the publicly accessible disk (one of
our disks is presently considered private because of this
restriction).

3. The extended virtual address space of the 2060 allows room
for the data tables that are
needed to describe the memory that we want to add to the system.
While it is possible to rearrange the operating system so that it can
have enough space to describe all the memory and all the disks, to
use memory in this way will cut us off from future software developments
that require this space. 

4. The upgrade to the 2060 seems necessary in the long run.  If it
is to be done at all, it should be concurrent with the memory upgrade
to be certain that the new memory is
compatible with the 2060's tighter timing tolerances.
.END

The price for the upgrade to the 2060 is $50,000, including the new
processor and a version of the operating system that supports the new
processor features and additional memory.  [All prices mentioned
herein are exclusive of sales tax, insurance, shipping, and
installation.]

THE MAJOR DIFFERENCE:  More memory

Analysis of operating data shows that in the worst cases observed,
only 30-40% of the CPU is delivered to users.  Another 20% goes to
scheduling users, and the remainder, sometimes more than 40% is
wasted.  The
problem of vast amounts of wasted time (technically known as "swap
wait") can be dealt with most effectively by adding more main memory.
We expect that with 512 Kw additional memory the swap wait time could
be dropped to 5-10% in the worst case, with a corresponding increase
in useful work to the 65-75% range. 

We have observed similar systems that have more main memory.  They
perform substantially better under heavy load than LOTS does.  Among
the systems we've seen are the DECsystem-1090T at SRI and the
KL10-based system at Artificial Intelligence Laboratory.  SRI has
1024 Kw of main memory, and runs essentially the same hardware and
software as LOTS.  Their system runs 100 interactive users and
performs much better than ours does when we are running 60 users.
The principal difference is that SRI has twice as much memory as we have.
The Artificial Intelligence Laboratory has recently added 512
Kw to their 384 Kw system.  Performance is now substantially better
than it was. 

There are two vendors of add-on memory.  DEC has recently announced a
solid-state memory for the DECsystem-20.  To double the existing
memory, using DEC's new memory will cost approximately $196,000. 
This cost breaks down as follows: 

.BEGIN INDENT 8,8; NOFILL;
756 Kw of MOS memory ∂(60)$250,000
256 Kw of core, trade in ∂(60)( 54,000)
Total:∂(60)$196,000
.END

The present memory is 512 Kw of core, in two 256 Kw boxes.  To make
room for the new memory, one box has to be removed.  The trade-in
allowance on the old memory is $54,000.  The new memory will be 768
Kw: 256 Kw replacing the core that was removed, plus 512 Kw
additional memory.  MOS memory from DEC cannot be delivered before
January 1979.  MOS memory absolutely requires the upgrade to the 2060
processor. 

A second plan also involves DEC as the vendor, but is unorthodox.
This plan calls for us to buy 512 Kw additional core memory and
connect it to the existing system.  Engineers at DEC are uncertain
about whether such a connection would work, particularly on a 2060
processor.  However, DEC is willing to let us experiment and if the
experiment reveals problems we may return the memory.  Also, DEC has
agreed to let us lease the memory on a trial basis.  The exact
details are somewhat complicated but the proposal is roughly:

.BEGIN INDENT 8,8
512 Kw memory and cabinet, $140,000.  Lease, $7,000/month,
of which 100% is applicable to purchase.  Maintenance while
on lease is an additional charge.
.END

It is not necessary for us to upgrade to a 2060 processor before
installing this memory.  However, there is a chance
that the core memory will not work, or will work less well with the
2060 processor than with the 2050 we already have.
It would be prudent for us to try the extended memory configuration
with the 2060 before buying the memory. 

The third plan involves another vendor, Ampex.  They propose to add
core memory to the existing system.  The price from Ampex for 512 Kw
of their core memory is $115,000.  The Ampex memory can also be had
on a short-term lease, for $5,000 monthly, with $4,480/mo applicable
to the purchase of the memory.  The disadvantages of going with the
Ampex memory are:

.BEGIN INDENT 8,8
1. Potential hardware problems preventing the memory from
working reliably.  (It's important that we evaluate
the performance of the memory on a 2060 system.)

2. Difficulties in dealing with suppliers and maintenance
in a multi-vendor environment.
.END

Offsetting these negative considerations is the price.  Moreover,
engineers at Ampex have demonstrated that their memory will work,
while there is some doubt that the DEC core will work in our
configuration.  It currently looks as though going with Ampex is
the best thing to do. 

ALSO NEEDED:  A second tape drive.

The present tape drive is adequate for most purposes, but when LOTS
is very busy it is difficult for a student to use the drive because
it is tied up, for many hours at a time, writing our disk backup
tapes.  Because of the general slowness of the system, we cannot
effectively backup all three disk packs; this prevents us from taking
full advantage of the third disk.  The tape drive is also our major
communication path to DEC, SRI (with whom we share software) and the
Graduate School of Business.  For our users, the tape drive is the
main path to SCIP.  We are anxious that people be able to take
advantage of both LOTS and SCIP, in whatever mix best suits their
needs.  Lack of an available tape drive inhibits people from
utilizing the best features of each system. 

To some extent, having more main memory is expected to relieve some
contention for the tape drive by making the disk backups go faster.
However, if we also expand to three public disks, we'll be 
back in the same place as before, with respect to accessing the tape
drive.

The additional tape drive costs $15,000.


IN SUMMARY:

The 2060 upgrade is necessary in the long run
to be compatible with DEC software developments and to take advantage
of new features as they become available.
However, if we buy memory, we need the processor upgrade concurrently, to
be sure the memory will work.

2060 Upgrade:∂(70)$ 50,000

Apart from an additional entire computer system, nothing will better help
LOTS meet the needs of the user community than more memory.  We need to
double the memory.  We have two offers meet the need by Fall quarter
1978.

512 Kw additional memory from DEC, purchase:∂(70)$140,000
.BEGIN INDENT 8,8;
short term lease: $7,000/month plus maintenance
100% of the lease price applicable to purchase
after 3 months.
.END

512 Kw additional memory from Ampex, purchase:∂(70)$115,000
.BEGIN INDENT 8,8;
short term lease: $5,000/month including
maintenance.  $4,480/month applicable to
the purchase for months 1 through 3.  For months
4 through 6, 60% of $4,480/mo is applicable to the
purchase.
.END

Purchase of an additional tape drive:∂(70)$ 15,000


Our recommendation:
.BEGIN INDENT 8,8

1.  Purchase of the upgrade to the 2060 processor∂(70)$ 50,000

2.  Purchase of an additional tape drive∂(70)$ 15,000

3.  Lease of an additional 512 Kw memory for 3 months,
including maintenance, not to exceed∂(70)$ 22,500

4.  If the memory and system behavior is satisfactory,
purchase of the memory, including allowance for
lease credits, not to exceed∂(70)$120,000

Total, exclusive of tax, shipping, insurance∂(70)$207,500

Estimated maintenance cost of the above,∂(70)$ 10,000/year
.END