Perspectives on cloud computing: interviews with five leading scientists from the cloud community

Gordon S. Blair, Fabio Kon, Walfredo Cirne, Dejan S. Milojicic, Raghu Ramakrishnan, Dan Reed, Dilma M. Da SilvaJ. Internet Serv. Appl 2011 · DOI

Abstract

Cloud computing is currently one of the major topics in distributed systems, with large numbers of papers being written on the topic, with major players in the industry releasing a range of software platforms offering novel Internet-based services and, most importantly, evidence of real impact on end user communities in terms of approaches to provisioning software services. Cloud computing though is at a formative stage, with a lot of hype surrounding the area, and this makes it difficult to see the true contribution and impact of the topic. Cloud computing is a central topic for the Journal of Internet Services and Applications (JISA) and indeed the most downloaded paper from the first year of JISA is concerned with the state-of-the-art and research challenges related to cloud computing. The Editors-in-Chief, Fabio Kon and Gordon Blair, therefore felt it was timely to seek clarification on the key issues around cloud computing and hence invited five leading scientists from industrial organizations central to cloud computing to answer a series of questions on the topic.

Introduction: Defining the Cloud Beyond the Hype

In the early 2010s, "cloud computing" was a term saturated with marketing hype, making it difficult for researchers and engineers to discern its true technical contribution. This paper provides a high-level "state of the union" for cloud computing circa 2011, featuring insights from top scientists at Google, HP, Yahoo!, Microsoft, and IBM. It clarifies that while the underlying technologies aren't necessarily new, the unprecedented convergence of scale, "rent-vs-buy" economics, and self-service elasticity represents a fundamental shift in how the world processes information.

Methodology: Structured Qualitative Interviews

Rather than a traditional experimental study, this paper uses a structured qualitative interview format. The Editors-in-Chief of the Journal of Internet Services and Applications (JISA) posed nine specific questions to five "cloud luminaries": Walfredo Cirne (Google), Dejan Milojicic (HP), Raghu Ramakrishnan (Yahoo!), Dan Reed (Microsoft), and Dilma Silva (IBM). Their responses were then synthesized to identify areas of consensus and divergence.

Key Insights and Ideas

The experts highlighted several core shifts that define the cloud era. Dan Reed (Microsoft) argued that cloud isn't about new technology, but about the shift in the ratios of capacity, capability, and cost. This shift enabled a dramatic improvement in the economics of massive-scale data centers. Most experts agreed that technologies like virtualization, HTTP, and distributed tables existed for years; the "aha moment" was the maturity of the business need and the timing of the deployment.

Dilma Silva (IBM) highlighted the democratization of high-end computing through cloud services.

Dilma Silva (IBM) emphasized that the cloud’s greatest impact is the "democratization" of high-end computing, allowing small startups and research groups to throw massive resources at a problem for a limited time without capital investment. This radical reduction in time-to-market allows innovation to flourish without the burden of physical infrastructure.

Technical Pillars and the Interoperability Bottleneck

Despite working for competitors, the scientists agreed on four pillars of cloud: (a) elasticity by design, (b) self-service, (c) multi-tenancy, and (d) pay-per-use. Regarding implementation, while the "SOAP vs. REST" debate was ongoing, the experts leaned toward RESTful APIs for their productivity and simplicity. However, a major hurdle remains: interoperability. Dejan Milojicic (HP) compared the 2011 cloud landscape to the early days of networking. Just as disparate networks eventually coalesced into the Internet via common protocols, the cloud requires a similar "IP moment" to overcome provider lock-in and enable a true "Federation of Clouds."

Limitations and Future Work

The transition to the cloud is not without its barriers. Walfredo Cirne (Google) noted a significant limitation: technical arguments may never be enough to convince highly sensitive organizations (like the CIA) to store their databases in a public cloud. Some applications will always remain in-house due to non-technical, psychological, or regulatory reasons.

Walfredo Cirne (Google) discussed the limitations of cloud adoption in sensitive sectors.

Furthermore, the experts admitted that we haven't yet abstracted "non-functional" requirements, such as guaranteed real-time latency or fine-grained privacy, into the generic cloud layer. Dan Reed pointed out that cloud computing needs to move from a "craft" to a "discipline" within university curricula, requiring more formal research into programming models that handle constant hardware failure at scale.

Conclusion: The Hybrid Reality

There is a strong consensus that the future isn't purely public or private. Large organizations will likely "flirt" with public clouds while keeping core business logic on private infrastructure, leading to a dominant "Hybrid" model. While the field is still in its formative stages, the shift toward elastic, utility-based computing is undeniable.

Summary updated

BibTeX
@article{blair2011perspectives,
  author = {Gordon S. Blair and Fabio Kon and Walfredo Cirne and Dejan S. Milojicic and Raghu Ramakrishnan and Dan Reed and Dilma M. Da Silva},
  title = {{Perspectives on cloud computing: interviews with five leading scientists from the cloud community}},
  journal = {J. Internet Serv. Appl.},
  year = {2011},
  doi = {10.1007/s13174-011-0023-1},
}
Figure