I can't confirm that.
On my cheap MacBook it takes some five minutes:


20:20 ~ $ time python3 -E -Wd -m test -r -w -uall test_mmap
Using random seed 1490092
[1/1] test_mmap
1 test OK.
[91067 refs]

real	4m50.301s
user	0m0.301s
sys	0m13.232s

...

20:21 ~ $ ll tmp/test_python_478/
total 6291456
 6291456 -rw-r-----  1 steffen  staff  6442450944  6 Apr 20:21 @test_478_tmp

...

Processes: 63 total, 2 running, 2 stuck, 59 sleeping, 251 threads                20:23:00
Load Avg: 0.85, 0.51, 0.24  CPU usage: 9.13% user, 15.38% sys, 75.48% idle
SharedLibs: 8260K resident, 9972K data, 0B linkedit.
MemRegions: 5563 total, 184M resident, 12M private, 391M shared.
PhysMem: 437M wired, 267M active, 113M inactive, 818M used, 1230M free.
VM: 140G vsize, 1042M framework vsize, 24419(0) pageins, 0(0) pageouts.
Networks: packets: 91/17K in, 101/17K out. Disks: 11128/551M read, 11089/8215M written.

PID   COMMAND      %CPU TIME     #TH  #WQ  #POR #MRE RPRVT  RSHRD  RSIZE  VPRVT  VSIZE
478   python3      4.0  00:07.22 2    0    37   137  13M    244K   15M    30M    2402M

...

20:24 ~ $ ll tmp/test_python_478/
total 5505024
 5505024 -rw-r-----  1 steffen  staff  5637144576  6 Apr 20:24 @test_478_tmp

...

Processes: 60 total, 2 running, 1 stuck, 57 sleeping, 246 threads                20:24:00
Load Avg: 1.28, 0.69, 0.33  CPU usage: 8.29% user, 16.9% sys, 75.60% idle
SharedLibs: 8260K resident, 9972K data, 0B linkedit.
MemRegions: 5444 total, 181M resident, 11M private, 535M shared.
PhysMem: 437M wired, 259M active, 124M inactive, 820M used, 1228M free.
VM: 133G vsize, 1042M framework vsize, 24421(0) pageins, 0(0) pageouts.
Networks: packets: 91/17K in, 101/17K out. Disks: 11128/551M read, 14697/11G written.

PID   COMMAND      %CPU TIME     #TH  #WQ  #POR #MRE RPRVT  RSHRD  RSIZE  VPRVT  VSIZE
478   python3      4.4  00:10.04 2    0    37   137  13M    244K   15M    30M    2402M


Is this the bot for which you've tracked those random failures?
Maybe this is a hardware failure, then?
(In the past i have had problems with my old PC and it's Elitegroup
K7S5A motherboard, running on FreeBSD 5.3.
That thing randomly produced "Stray IRQ 11" (about ~handful a day),
and it's ATA controller (i think) randomly caused complete hard disk
regions to become unreadable until after a restart.)