In-Depth Analysis
Prof Derek Keats, deputy vice chancellor of Wits University, is building private cloud infrastructure for the university with application software developed in Africa.
African free software stalwart Prof Derek Keats is propelling Wits University into the cloud as an efficient way of managing resources at the institution.
One of the first things you notice about professor Derek Keats is his impatience. He’s impatient with national initiatives that he says have failed to move out of the ‘talk shop’ and into ‘developing sensible infrastructure’; he’s impatient with the way people treat open source software as an untrustworthy third cousin of proprietary models, and he’s impatient with being a mere user of technology developed elsewhere. This makes him the perfect candidate for pioneering private cloud computing at Wits as the new deputy vice chancellor of Knowledge and Information Management.
To his knowledge, this is the first enterprise implementation of cloud technologies at a South African university. He believes that bringing together computing resources in this way will enable digital preservation and archiving services previously unavailable to the university, as well as being able to better support academic research initiatives.
Although he is reticent about claiming that the initiative is ‘unique’, he believes that “no other institution has used synergy within its core enterprise infrastructure to create archiving capability that cuts across all aspects of archiving and preservation and does it in such a way that it could easily tie into national initiatives”.
Keats says that cloud architecture enables Wits to do more with the same resources.
“We are using a cloud architecture to create synergy among various initiatives,” he says. “With funding for e-mail, web and eLearning, for example, we are also able to create the infrastructure for digital archiving and preservation, enterprise document management and, later this year, we will be able to provide for simple and easy hosting of virtual servers for academic and research use.”
The eLearning and web infrastructure cloud services at Wits are both using applications made in Africa as part of African Virtual Open Initiatives and Resources (AVOIR), a free software project that Keats has been driving since inception. AVOIR is a network of 16 African universities with partners in North America, Europe and Afghanistan that was initiated to build capacity in local software engineering. Although driven primarily by the University of the Western Cape, where Keats worked for 20 years before he joined Wits last year, the network is growing steadily, with Wits and the University of Mauritius the latest in a diverse range of members.
AVOIR is billed by Keats as a pan-African network for building capacity in software engineering. In building such capacity, AVOIR has developed Chisimba, a Web 2.0-enabled application development framework that runs some of the cloud services that Wits is using for this initiative.
Moving to the cloud In moving to the cloud, Wits University is following a global trend that is driving cloud computing in business, government and academia. Pundits believe that economics is driving the move to the cloud.
By sharing computing power among multiple tenants, cloud computing can improve utilisation rates as servers are not unnecessarily left idle, leading to faster, simpler and cheaper use of infrastructure.
According to Nicholas Carr, the outspoken American technology writer who, in 2004, argued that the strategic importance of information technology is diminishing as it becomes standardised and less expensive, the cloud computing paradigm shift is similar to the displacement of electricity generators by electricity grids early in the 20th century.
His book The Big Switch describes how companies stopped generating their own power with steam engines and dynamos when the electric grid was built 100 years ago. “The cheap power pumped out by electric utilities didn’t just change how businesses operated. It set off a chain reaction of economic and social transformations that brought the modern world into existence,” he writes.
“Today, a similar revolution is under way. Hooked up to the internet’s global computing grid, massive information-processing plants have begun pumping data and software code into our homes and businesses. This time, it’s computing that’s turning into a utility.”
Others are more sceptical, claiming that cost savings are not always significant and that cloud computing increases the dependency on a few software vendors. In Wits’ case, the initiative seems to be dependent on the success of the Chisimba platform, which, in turn, is dependent on the ongoing involvement of collaborating partners.
Keats says that vendor dependency is not an issue because this is a private cloud that is being built with free software. “There is no lock-in and we could change the software technology and rebuild the stack very quickly should we need to,” he states.
Keats believes that there are obvious incentives for institutions to collaborate with one another in developing Chisimba. “There is huge potential for institutions to collaborate in both developing cloud-based applications such as we have done with Chisimba, but also, importantly, investing in shared hosting facilities,” he adds.
Although the cloud itself will be private to Wits, collaboration seems to be key to this initiative. According to Keats: “This is a Wits internal initiative, but we have benefited enormously from participation in the Preservation and Archiving Special Interest Group organised by Art Pasquinelli from Sun Microsystems and Mike Keller from Stanford University.”
The project has not been without its challenges. “Working with Sun, there has been some chaos around the purchase of Sun by Oracle,” says Keats. “Local skills are also lacking, but we will soon have those based on this experience.”
Interestingly, Keats notes that the biggest challenge is having enough electricity to supply Wits’ data centre.
Nicholas Carr’s analogy with the electric grid is pretty ironic here. As South Africa tries to innovate with the rest of the world in developing new computing utilities, we’re saddled with an older infrastructure problem that needs a solution at the same time.
If anyone is going to surge ahead and face these challenges head-on, it’s going to be Derek Keats. His impatience, in this case, is the virtue here.
