I thank you for doing this! Long ago and far away I did report a problem with
SWAPGEN. I took this up with IBM, since they were providing our Linux (Red Hat)
support and since they used SWAPGEN in their RedBook. Eventually, IBM
determined that the problem was not in SWAPGEN but in systemd. They
Be nice to be available a vdso enabled syscall like gettimeofday.
Original message
From: Timothy Sipples
Date: 1/9/19 19:29 (GMT-05:00)
To: LINUX-390@VM.MARIST.EDU
Subject: Re: [LINUX-390] VoltDB on z?
Christian Borntraeger wrote:
>We do have both variants (crc32 and crc32c)
Christian Borntraeger wrote:
>We do have both variants (crc32 and crc32c) as vector code in
>the kernel
Yes, this support was introduced in Linux kernel 4.8 which was released in
October, 2016. It requires a z13 processor or higher. It's perhaps possible
a distributor backported this feature to a
z14's are 5.0 GHz...
Lee Stewart ● VM System Support ● Visa ● Phone: 6(750)4601 - +1-303-389-4601 ●
lstew...@visa.com
-Original Message-
From: Linux on 390 Port On Behalf Of Neale Ferguson
Sent: Wednesday, January 09, 2019 11:04 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: VoltDB on z?
I know some government locations have been looking at this and that’s what
really started my thoughts on posting.
And yes, while some intel servers are up over 3ghz and can be overclocked,
that may not be normal from the cloud based services that are today’s real
competition.
On Wed, Jan 9, 2019
For this particular package there is a good chance Z would be a good choice:
- Data in memory: where Z does shine. If huge page support is available then
even better. So much of the Z chip real estate is devoted to caching, TLBs etc.
- Pause-less garbage collector using the IBM JDK (the engine
Kurt wrote:
>5.2 Ghz compared to not quite 2 Ghz.
>Same reason IBM shows most open source packages now run 40% faster (or with
>40% more throughput) then similarly defined (core and memory) x86
>environments in studies.
>That of course does not really tell the whole story. Offloading i/o and
>the