The Parallel Universes of Cloud Computing
Science fiction mainstay helps explain some common but divergent views on cloud computing.
Parallel universes are staples of science fiction writing. They may also be the only explanation for some common but divergent views on cloud computing.
In one case you have the good cloud universe where IaaS is the foundation of billion dollar valuations and ground breaking science. But in its doppleganger, the bad cloud universe, IaaS is a money sucking vortex that will leave you a penniless shadow of the IT professional you once were.
Well, after I’ve spent a week or so installing Linux, NGINX, Varnish/Memcache, PHP, WordPress etc. Then spend time hardening the Linux, NGINX, WordPress etc. And configuring the email server. And the DNS server. And patch/update it every week or two. And after all that, my hosting costs spiraled out of control almost immediately. It was a complete failure from start to finish.
Meanwhile, from the good cloud universe we get a steady stream of positive reports ranging from advances in science:
That software bundle performed a daunting simulation of 21 million chemical compounds to see if they would bind with a protein target. That run ate up the equivalent of 12.5 processor years, but it was completed in less than three hours.
The trick for the rest of us is to figure out if we inhabit the good cloud universe or the bad cloud universe. Here are some clues to help you out.
Occupants of the good cloud universe are often starting a new company or product offering. They don’t want to spend your limited capital on hardware so they go with a public IaaS solution. They are also prepared for substantial networking charges if you users will be downloading large volumes of data.
Others from the good cloud universe have a lot of variability in their need for computation. It makes no sense for them to invest in a large cluster that sits idle most of the time. They would rather rent out a cluster on Amazon, run their large Hadoop jobs and then shut down their instances. This works well and is why we will keep hearing stories about science in the cloud.
A different set of requirements plague inhabitants of the bad cloud universe. They have a relatively constant level of demand for Web applications. They tend to be better off with a traditional hosted solution where you pay a fixed amount for a service. Cloud pricing schemes are designed for dynamic use; if that doesn’t fit your needs then don’t use it.
A tell-tale sign you and your application are in the bad cloud universe is lack of control. If you run your application in the cloud but can’t control the demand for the services you are charged for then your days in business are numbered. Amazon charges for data transfer out and it doesn’t care who is downloading. At EtherialMind.com they found:
...the Web sites are being scraped on the RSS feeds from hundreds of separate sources including “market sentiment analysis” companies, chinese scrapers, internal corporate intranets, vendors feeds and much more that could not be identified.
In another case of out of control charges, an NYU professor run up a $1,200 bill in a few hours conducting a crowd sourcing experiment which he later dubbed a “denial of money” attack.
Watch out for crosstalk between the universes. That happens when people from one cloud universe assumes their requirements apply to people in the other universe. It can take the form of simplistic economics arguments along the lines of “Amazon has economies of scale you can never realize, you couldn’t possibly do it for less.” Or it can take the “you are missing something” line of reasoning. Greg Ferro at EtherialMind.com got that kind of response to his post:
A few people attempted to make the point that I didn’t understand, or didn’t know something or other.
Judging from his posts, I doubt he lacks understanding.
It pays to know which universe you are in before putting your ops into a public IaaS cloud.
Dan Sullivan is an author, systems architect, and consultant with over 20 years of IT experience with engagements in systems architecture, enterprise security, advanced analytics and business intelligence. He has worked in a broad range of industries, including financial services, manufacturing, pharmaceuticals, software development, government, retail, gas and oil production, power generation, life sciences, and education. Dan has written 16 books and numerous articles and white papers about topics ranging from data warehousing, Cloud Computing and advanced analytics to security management, collaboration, and text mining.
See here for all of Dan's Tom's IT Pro articles.