> On Thursday, August 3, 2023 at 10:47:52 AM PDT, Joel C. Ewing > <jce.ebe...@cox.net> wrote:
> There is a synergy that exists between z-architecture hardware and z/OS > that has evolved over many decades. IBM designs with insight whereas other manufacturers implement. You would never install 1 giant disk on IBM z. Instead you install multiple smaller disks to avoid bottlenecks. There is nothing that stops other manufacturers from designing everything IBM has designed. IBM thinks about everything from RAS to performance. > The hardware is designed with redundancy to detect failures There is no doubt that IBM does an excellent job in RAS design. You get what you pay for. Don't expect to get a million $ computer for $5,000. > Undetected hardware errors don't happen. Undetected by nature goes unnoticed. It's extremely rare for IBM but People aren't perfect and solutions to a problem simply change the problem. > designed with the philosophy that software failures may occur within > parts of the operating system, either from a hardware failure or a > system software bug. System recovery routines exist to clean up after > such failures, limit what running address spaces are affected, and allow > production to continue in unaffected address spaces. While IBM goes to great strides for RAS, they have their limits. GDPS and automation costs extra. They exist purely for RAS. > An explicit part of the design philosophy is that applications running > in different address spaces are isolated: CICS and z/OS UNIX use an address space to run multiple applications (at least they used to). I suspect that programming languages have become so reliable that we simply don't experience many storage overlays caused by applications. On the other hand, UNIX and Linux processes are fully protected with each process having storage protected from other processes. > Another important feature of z/OS that requires some hardware > coordination is the System Measurement Facility that gathers measurement > of system activity and resource usage at a level to support performance > tuning or billing based on resource usage. SMF and RMF are simply free tools that provide basic information. Unix and Linux also provide a similar free tool for monitoring performance. On z/OS, you can upgrade to Omegamon, Intune and other tools for more in depth performance information. > if you could somehow succeed in running it under > Linux or on non-z hardware, it would lose the reliability, availability, > and serviceability it gets from that hardware/software synergy that > makes it an ideal production platform for critical workloads. What I'm saying is that we should expect some z/OS components (not all) to appear in IBM RHEL on IBM z machines. Don't expect CICS on RHEL for z but migrating GRS is minor considering that the hardware and instruction set are the same. You only need to change the z/OS specific code. GRS improves RAS because it's a superior implementation of Unix mutex and semaphore. Same for SYSPLEX, some of DFSMS, JES, SSI and more. To have the same synergy as z/OS, RHEL only needs the components that provide the synergy. I've never said that z/OS will be implemented in non-z hardware. I've said there is nothing stopping other companies from implementing these designs. They simply have no desire to use 2 PCIe ports for redundancy when 1 works most of the time. I've never said that IBM would integrate z/OS into Linux but said in theory they have prior experience in doing such things like this. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
