Comment: Cloud empowers scientists to accelerate innovation

Contributed by Shane Owenby – Managing Director, Asia Pacific, Amazon Web Services

It’s the year 2000 and a research group at a biotechnology company is discussing an edgy and ambitious whole genome RNA sequencing project. But a project of this magnitude will produce terabytes of data and need vast computing power to be used in an undefined number of permutations and combinations.

It’s clear the technology firepower needed to get this project off the ground is beyond the group’s budgetary means. Progress stalls. Scientific thresholds remain just that – thresholds.

The scenario in 2011 is very different. IT procurement, costs and lengthy technical reviews once cast long shadows over research groups, but cloud computing has virtually removed resource constraints as a barrier to progress.

Using cloud computing, rather than buying, owning and maintaining their own data centres or servers, organisations purchase infrastructure resources such as computing power and storage services from third-party cloud providers on an as-needed basis.

Database, messaging infrastructure and content distribution services can also live in the cloud. The provider manages and maintains the entire infrastructure in a secure environment and users interact with their resources via the internet. Capacity can grow or shrink instantly and users pay for what they actually use.

With a cloud computing strategy in hand the biotechnology company’s research group, stymied in 2000, likely could fund their whole project out of a discretionary budget. It can access on-demand technology infrastructure with enabling technologies such as scalable storage, elastic compute power and dynamic analysis platforms with no upfront cost or negotiations.

And so they begin by sending samples to a sequencing service provider, who ships the results to a secure cloud environment. The storage needed for the results is available on-demand with a pay-as-you-go pricing model, meaning the researchers pay nothing until the first byte is written or after the final file is removed.

The collaborators get straight to work performing large-scale, distributed computations. Sharing results becomes as easy as sending an email. Their research program, at last, is enabled to run free of cost and infrastructure restraints. The thresholds can now be crossed.

The availability of cloud computing has clearly changed the way many organisations acquire IT, but for scientists the changes run far deeper. Researchers in industry and academia are using computers in continuously increasing quantities for molecular simulation, virtual screening and DNA and protein sequence analysis, generating vast quantities of data.

In the past, organisations had only two options: spend big on expensive, purpose-built cluster resources and data management systems with high associated management costs; or use shared infrastructure, often at supercomputing centres, and be forced to wait for access, often for weeks or even months.

Both scenarios channelled away the precious resources of time and money from the real task at hand: research. Additionally, the progression and scale of scientific research and development and the demands it places on resources are often unpredictable.

In six months, a project’s technology requirements may have changed three to four times (or more). The affordability, scalability and accessibility of cloud computing is invaluable.

• Share
  Share this article

  •  google+
  • facebookfacebook
  • slashdotslashdot
  • diggdigg
  • RedditReddit
  • stumbleuponstumbleupon
  • linkedinlinkedin
  • twittertwitter
• print
• email
• Bookmark