[Info-vax] What would you miss if DECnet got the chop? Was: "bad select 38" (OpenSSL on VMS)

[Kerry Main]

> -----Original Message-----
> From: Info-vax [mailto:info-vax-bounces at rbnsn.com] On Behalf
> Of Michael Moroney via Info-vax
> Sent: 07-Oct-16 2:11 AM
> To: info-vax at rbnsn.com
> Cc: Michael Moroney <moroney at world.std.spaamtrap.com>
> Subject: Re: [Info-vax] What would you miss if DECnet got the
> chop? Was: "bad select 38" (OpenSSL on VMS)
> 
> Stephen Hoffman <seaohveh at hoffmanlabs.invalid> writes:
> 

[snip..]

> >Oddly, the rest of the universe gets by with ssh, netcat, file
> shares
> >and related.
> 
> Yes and the rest of the world gets by without shared-everything
> clusters and M$ PCs filled with bloatware that have to be
> rebooted every few days.
> 

For a real world example of how the rest of the world "gets by"
without shared everything clusters, I would encourage watching
this video presentation by a fairly well known distributed
systems engineer who now works for Twitter, but previously worked
for Microsoft. She describes the issues and her views of
"solutions"(??) on how to maintain these distributed systems at
the application code level (shared db's, replication, load
balancing and lack of data consistency) etc.

https://www.youtube.com/watch?v=H0i_bXKwujQ
"Building Scalable System Services" - Catie McCaffrey

As you watch this video, listen closely to the mental gymnastics
these distributed systems developers need to deal with these
challenges.

Big difference is that the shared nothing world maintains data
consistency, node mgmt. (adds, deletes), HA, DR etc at the
Application code level while all of this is maintained in the
shared everything (shared disk) environment at the OS level.
Think about this - the app coders need to know and deal with node
failures, adds in their APPLICATION code.

 [snip]


Regards,

Kerry Main
Kerry dot main at starkgaming dot com