[Discuss] Whence distributed operating systems?

[darose at darose.net (David Rosenstrauch)]

On 04/21/2016 12:38 PM, Mike Small wrote:
> David Rosenstrauch <darose at darose.net> writes:
>
>> On 04/21/2016 12:50 AM, Mike Small wrote:
>>> "Sadly it seems that we now need to either wait for Linux or Windows to
>>> catch up with the 1980s state of the art in distributed systems (think
>>> Locus or AFS). What went wrong? Products like DataSynapse?s FabricServer
>>> look like an interesting attempt to address the problem, at least for
>>> the Java world, but it feels to me that mainstream operating systems
>>> designers seem to have lost the plot somewhere along the way."
>>>
>>> http://discovery.bmc.com/community/blog-post/whatever-happened-to-distributed-operating-systems3/
>>>
>>> Is single system image still a thing?
>>
>> Aren't systems like Apache Mesos (which didn't exist back nearly 10
>> years ago in 2007 when the author wrote that post) the natural successor
>> to DataSynapse FabricServer, and an example of the "distibuted operating
>> system" he's talking about?  I.e., just a big pool of CPU cores, where
>> different portions of the pool can be utilized for different types of
>> distributed workloads.
>
> Sounds more like what he's talking about. Are these kinds of systems
> gaining much traction? I'd never heard of Mesos.

Yes, I think so.  Many companies are starting to migrate their 
Hadoop/Spark/etc. infrastructure on to it.  (We're starting to do that 
here at YP.)

It makes sense:  if you're using Hadoop, Spark, and other distributed 
systems, it doesn't take long to figure out it's more efficient to have 
that a system that can run any of those on the same cluster simultaneously.

Mesos came out of the same Berkeley Lab that built Spark.  More info on 
its history here:  https://en.wikipedia.org/wiki/Apache_Mesos

DR