PC slower with cache !!

PC slower with cache !!

Post by Andrew Davi » Sat, 27 Sep 1997 04:00:00

My PC is slower with cache added ! What's happening ?

Whoops, nearly hit the send key ... just kidding :-)

It's an old Packard Bell 486ES. I got 5x W24257AK-15, which I gather
are 15ns 32kb chips. The manual says I need 25ns chips for data and
one 20ns chip for tag, so I figure that 15ns chips are OK. When I add
the chips, the BIOS says "128Kb cache". However, when I run a benchmark,
things are much slower with the cache. From the BYTE Unix benchmarks,
version 3.11:

                         | ---  no cache -|    |- 128kb cache -|
                         486-33     586-133    486-33    586-133  (P90)    (486-66)
Dhrystone 2        lps   19968      45590      3255      7527     (80123)  (35314)
(without reg. var.)
C Compiler         lpm   16.1       39.4       5.9       14.6     (74)     (48)
sqrt(2)            lpm   1617       1132       377       205      (9056)   (2438)
Hanoi              lps   401        466        115       250      (1572)   (812)

I got a Gainbery GB586-133 overdrive processor too. The main memory is 28Mb of 70ns
DRAM. The figures in brackets  are
for other systems - not sure what the cache is on them.

So, for now, the cache chips are back in the drawer.

It seems weird that things should be slower with the cache in; I was wondering if
something's broken. If I misconfigure the jumpers, the computer won't boot
properly, so I figure they're OK.


Richmond BC, Canada   fax:604-270-8285


1. 2.5 slower with caching?

: I've had a few users who are running/testing Solaris 2.5 with caching.
: They seem to feel that it runs "slow". Not very precise in what is slow,
: but when I turned off caching they reported that it was "faster" or at
: least more consistent.  Has anyone else heard these types of complaints?
: I am caching only application directories (framemaker, etc.) and they
: are running on a Sparc 5/75 and IPXes.  Surely accessing from local disk
: should be faster than going over the network to our server even if I
: have it consistency checking every 30 seconds?

not in every case.

cacheos, in my experience, isn't something that's always useful,
or to be blindly used. it's just like any other tool or component
of an overall system design.

for example, I had several machines here set up with a cache for
their software directories. everyone noticed that those machines
were significantly slower than all the others. didn't make anyone
happy, that. why? because the machine had only one disk. as a result,
when it started paging, it'd flush out both anon memory and software
text... which it'd then have to reload - from the cache.. and in this
case, all the extra overhead of moving the disk heads back and forth
from swap partition to cache partition and back again and again was
overall slower than reloading the software from the relatively faster
server and just swapping to the local disk.

: One other aspect is that with caching on (30 second timeout to test it),
: when they type a command that doesn't exist in their PATH, it takes 10
: or 15 seconds to return for the first mistyped command.  Then its
: instantaneous for a few hours, then it takes 10 or 15 seconds, etc.

turn off consistency checking. :)

= Jeff Bacon         General Systems Hack, Michigan Technological University =

2. Goto Pro's and Con's

3. Q: System w/ limited 256 kB cache, 16MB->32MB = slower linux.

4. kwintv: How do I know if I have Video4Linux correct? (RH 6.0 - KDE 1.1.1)

5. squid web cache: the larger the slower

6. KDevelop, sub folder with source files, make problem

7. Mpeg3

8. PPP gets slower and slower

9. telnet logins running slower and slower...

10. Server getting slower and slower. Why?

11. Workshop 4.2 -> 5.0 -> 6.0 slower and slower...

12. logout gets slower and slower!