Today's advent of distributed, massive-scale "cloud computing" is something of a return to the early 1980s, when computing was of a different sort. Rather than individual desktop or laptop machines, which are the current norm, workplace computers were usually time-shared among multiple users working on "dumb" terminals connected to a central machine--often located in some remote corner of the building. According to the New York Times, at least one major pharmaceutical company, Eli Lilly, already is using cloud computing to support its R&D efforts. Lilly uses Amazon's cloud services for some R&D. activities. There are definitely concerns about data security. Dave Powers, an Eli Lilly senior systems engineer, noted that they are very careful about what information is putting on the Amazon servers.
Cloud computing may offer the greatest benefit to small laboratories with limited information technology budgets. Already about 3.2% of U.S. small businesses, about 230,000 companies, use cloud computing services. Another 3.6%, 260,000 businesses, plan to add cloud services in the next 12 months. Spending on cloud computing by small business will increase by 36.2% from 2009 to US$2.4 billion.
What is cloud computing
"Cloud computing" is a term that confuses many lab managers, scientists and research engineers. In a sense cloud computing is back to the future, a return to the early days of computing in the 1980s. Cloud computing enables your workstation to connect to remote, massive, warehouse-scale data centers housing large networks of processors and memory. This capacity is used to process and store data. These big datacenters offer economy of scale that lowers computing costs according to network architect Cedric Lam of Google.
An outside vendor such as Amazon or Google runs the remote servers and software. Lab managers can have smaller information technology departments in-house and focus on the lab's primary business: meeting the technical service needs of the firm's customers and developing new products and processes.
Another advantage is that these datacenters relieve individual laboratories of the hassles of maintaining and upgrading equipment. Also, despite many admonishments to do so, some laboratory staff members fail to back up their data.
Perhaps the biggest real advantage of cloud computing is portability, suggests Lam. He notes that with a network connection users can access their data from anywhere and do so at any time.
Improving the state of the art
Currently data centers are huge power hogs. However, Lam believes that use of low-cost, high bandwidth/high-density optical fiber interconnects will reduce power consumption while increasing information carrying capacity. Increasingly optical cables are used for long-distance connectivity between datacenters and between users and datacenters.
Some laboratory managers may be concerned about the security of proprietary data when using cloud computing and storing data and data processing protocols in a distant data center. Even if the lab manager is comfortable doing so, will his colleagues in other units of the company accept of cloud computing? What about the company's customers? Security of data, particularly patient's names, is essential in drug clinical trials.
Upfront costs for cloud computing are relatively moderate since the laboratory needs little hardware and relatively few information technology staff members. However, cost reductions touted for cloud computing may not be applicable for firms with large research centers already processing high volumes of data. In some of these cases, the situation is akin to deciding whether to buy or lease a car.