As part of ZE’s continuing commitment to stay abreast of new Information System trends, Yi-Jeng recently attended Bio-IT World Asia, held June 6-8, 2012 in Singapore. The conference brought together IT professionals, bioinformaticians, pharmaceutical and biomedical scientists from all over the world to discuss the increasing intertwined field of IT and Life Sciences.

Academic institutions and bioinformaticians present at BIO-IT Asia 2012 were interested in solutions for two classic big-data problems:

  1. How to compress large quantities of genomic data without loss of fidelity?
  2. How to create IT infrastructures that could scale up quickly to service unpredictable computational and storage demands?

The solution to the first question required innovation by bioinformaticians and computational biologists, which is outside of the scope of this blog. However, as they innovate and compress biological data, this compression invariably increases the computational load; so in reality, storage and computational issues are linked.

The second question addresses the need for IT infrastructure to react quickly to increasing computational needs. To meet this challenge, many institutions present at the conference are creating private cloud computing services for their constituents; with the option to tap commercial cloud computing services during peak usage.

Figure 3. Cloud-Expandable IT Infrastructure at National Bioinformatics Institutions

Figure 3 shows two large bioinformatics centers, Australia’s Victorian Life Sciences Computation Initiative, and South Korea’s Theragen Bio Institute. On the surface, they couldn’t be more different from a business model or even a cultural perspective. Nevertheless both institutes saw value tapping commercial cloud computing services (I added Red “Cloud” arrows for emphasis).

Pros and Cons of Cloud Computing

A cloud computing discussion often attracts legions of fans and distractors alike. Proponents argue clouds allow ‘cost-saving through flexibility’; a concept worth exploring. Traditional IT infrastructure projects requires accurate forecast of usage demand for a future time horizon, which could lead to costly up-front IT investments without a proven demand. Clouds, argues proponents, allows ‘flexibility’ to ditch demand forecasting (and potential waste), and invest in IT only when demand is proven in the marketplace. Amazon Web Services’ (AWS) diagram in Figure 4 offers a visualization of this concept.

Figure 4. According to Amazon Web Services, their main advantage is to help customers replace up-front capital IT expense with lower variable cost.

Yet for all the purported advantages in flexibility and cost savings, here is still a perception in industry that data dispersed in commercial clouds are less secure than data in private server rooms under company lock and key. Although AWS and other cloud companies incorporated the use of advanced asymmetric key cryptography (public/private keys), and is probably as secure (if not more so) as compared with other corporate networks connected to the internet, the perception of a security gap persists. In the open research, cash-poor world of academia, high adoption rates of AWS suggests that security is not a concern. In the corporate world however, where extensive customer and proprietary data exists, data security (perceived or real) is still a primary concern. It remains to be seen if cloud companies will be able to convince corporations to trust cloud security enough to take advantage ‘cost-saving through flexibility’. We at ZE continue to monitor the evolution of cloud computing and to evaluate potential applications for ZEMA. Currently our customers are telling us they would rather pay IT overheads and keep their data centers securely on-site; but given that the field of commercial cloud computing is still maturing, opinions may change in the future.

What is your organization’s commercial cloud computing strategy? Write to us, because we’d like to know!