Cloud Computing - Part 2.0 - Evolution (or just intelligent design?)
The first part of this entry was published on March 11 here. The saga continues.
In the 1980s, in the arena of big data processing users the cord ran from a workstation to a large mainframe or AS400 computer, which was often in the same room or in close proximity. The cord was whole, undivided and dedicated. Nothing would interfere with the communication between the workstation and the processor unless the cord was cut or the commands from the station caused the processor to work harder. Most of the computing assets were owned and maintained by the users and programming and maintenance staffs were huge.
The next step was removing the processor from the general vicinity of the workstation. The processor could be in the next building or the next county or state but was still connected by a dedicated cord (e.g. a dedicated T1 phone line or equivalent).
At some point in this process, it became advantageous for some users to not own the computing resources and “outsourcing” or “facilities management” came into vogue. With these delivery models, a party other than the user owned and controlled some part of the computing environment. However, both parties together owned and/or controlled the entirety of the environment with the exception of the lines leased from the phone company. Some companies even used the satellite transmission of data so the connecting “cord” traveled from the user to a satellite, down to the processing site and then in reverse.
Along with these changes, the processing environment became more diverse. Instead of connecting with one machine or one logical partition within the machine, data may be processed on numerous machines and stored on others and may be moved from one machine to another as needed by the processor and to optimize computing efficiency. At this point, the user could not directly identify the specific machine upon which its data resided or was processed, although the user generally knew the location of the data center or centers and could, if necessary, walk in the door and see and touch the machines where the data was processed and stored.
Cue the Internet.
At some point in this timeline, the internet intruded. This made it possible to use the internet as the connecting cord. Data was broken down and transmitted in little packets through the internet from the terminal to the processor (which was now probably a server rather than a hulking mainframe) and back. The cord was no longer dedicated and unbroken (or even necessarily a physical entity) but the user still generally knew the location and identity of the processing assets.
The inexorable march of Moore’s Law, the continual fall in the price of hardware and the tremendous success of online shopping sites like E-Bay and Amazon led to the realization that the amount of computing resources needed at Christmas and Valentine’s Day was far greater than that needed the rest of the year. So, how best to leverage the use of these vast server farms the rest of the year? Data management software (also sometimes called virtualization in this context) that allowed the identification and rerouting of data and processing resources led to the business model that allowed the owners of these server farms to rent out the use of the resources during non-peak times. Data (and applications) could be stored and installed on these servers and moved quickly and efficiently (hopefully) as the need arose.
Cue the cloud.
Workstations or laptops with a web browser and an internet connection could be used to process and transmit (or only transmit) data into the internet, which would then be directed to the appropriate server or servers, which would store and/or process the data and return to the user via the internet. The environment has evolved from a workstation with a cord connected directly to a processor in the same room to an environment where the user has no idea where the processor may reside and, in fact, it likely will not be the same processor or location for any discernable time. Computing power can be parceled out, distributed, repurposed and moved freely and without notice. It might be in the next room but could just as easily be on another continent. The “cloud” analogy came to be used because everything outside the workstation is amorphous, changing and fairly non-definable. No longer can a user easily walk into a facility and touch and feel the processor that is doing the work.
The most basic operational and legal issues surrounding the provision of these services have not changed appreciably over the above-described evolution. The user still wants to have an acceptable degree of comfort with response time, on-line up time, through put, security, business continuity and disaster recovery as well as help desk services and software and hardware updates and refreshes.
While the basic issues have not materially changed, the complexity of addressing them has. The constant variability of the environment and the fact that a large part of it is not controlled by either the provider or the user makes the addressing of these issues more difficult from a legal and an operational sense.
In the coming posts, we will address some of these issues as they pertain to models referred to as “software as a service”, platform as a service”, “infrastructure as a service” and yes, even “everything as a service”.
We may well find out we don’t know clouds at all. Cue the cheesy lyrics.
Reminded me of this quote:
Abe Simpson: "We can't bust heads like we used to. But we have our ways. One trick is to tell stories that don't go anywhere. Like the time I caught the ferry to Shelbyville. I needed a new heel for m'shoe. So I decided to go to Morganville, which is what they called Shelbyville in those days. So I tied an onion to my belt. Which was the style at the time. Now, to take the ferry cost a nickel, and in those days, nickels had pictures of bumblebees on 'em. Gimme five bees for a quarter, you'd say. Now where was I... oh yeah. The important thing was that I had an onion tied to my belt, which was the style at the time. You couldn't get white onions, because of the war. The only thing you could get was those big yellow ones..."
Basically a very good thing, has only 2 downsides:
1. privacy & security (you really must trust that system, to leave your files out there on the "world wilde web" with no real control over them)
2. environmental issues - to only mention one of the cloud computing promoters, it was estimated that Google’s data centres contain nearly a million servers, each drawing about 1 kilowatt of electricity and so every hour Google burns through 1 million kilowatt-hours. Imagine an increase in couding technology in the future! It will only translate in even more burned power ...