My Friend Alan Rocker and I often discuss ideas about technology and tradeoffs. Alan asked about SSDs for Linux:
> I haven't been following hardware developments very closely for a while, so I
> find it hard to judge the arguments. What's important?
Ultimately what's important is the management software, the layer above the drivers, off to one side. That applies regardless of the media and means that the view the applications take of storage is preserved regardless of changes in the physical media.
> The first question is, what areas are currently the bottlenecks and
> constraints, at what orders of magnitude?
The simple answer is 'channels'.
What's the saying "Those who forget history are doomed to repeat it over again"?
Weren't we doing this with routers and ... well if not firewalls as such then certainly filtering rules in the routers, way back in the 1980s?
I recall attending a luncheon put on by Dell about "Software Defined networking". Basically it was having routers that were 'agile' enough to change routing and implement tactical policy by load, demand and new devices or devices making processing demands.
Again we were doing that in the 1980s. Working with ANS as they cut over the academic internet to the commercial internet with their "CO+RE" pseudo-product. basically it was that they had been supporting the academic internet and were not selling commercial services using the same backbones, trunks and "outlets" (sometimes known as 'point of presence'). This 'policy based routing' was carried out by custom built routers; they were IBM AIX desktop boxes -- the kind I'd used to implement an Oracle based time management/billing system for at Public Works Ottawa a few years earlier, along with some custom built T3 interface cards.