CLIENTS

February Issue Reports on Symbolic Execution for Software Testing, Promoting Better Passwords, Cloud Service Certification, and Reflections on MOOCs

Delivering the next generation of cloud services requires consistently responsive massive-scale computing systems that keep the tail of latency distribution short, write Google Inc. Fellows Jeffrey Dean and Luiz André Barroso.   In the February cover story of Communications of the ACM cacm.acm.org/magazines/2013/2 , they outline some common causes for high-latency episodes in large online services and describe “tail-tolerant” techniques that reduce their severity or mitigate their effect on whole-system performance.   Since these techniques often rely on existing capacity for fault-tolerance, they can be implemented with a modest overhead. 

Also in this issue, ACM Publications Board Co-Chairs Ronald Boisvert and Jack Davidson address the critical role of non-profit professional societies in developing sustainable economic models in the age of open access.  They describe the array of options that ACM has chosen for ACM authors and Special interest Groups to enable a natural, stable evolution of the publication enterprise into the future.

In his column, ACM President Vint Cerf surveys the multidimensional efforts of the computing field, including ACM, to stimulate interest in all aspects of computing.  He challenges readers to submit their ideas for broadening the participation of a demographically diverse population.    

Communications, the flagship publication of ACM acm.org offers readers access to this generation’s most significant leaders and innovators in computing and information technology, and is available online in digital format.

            Other Communications highlights:

• Cristian Cadar of Imperial College London and Koushik Sen of the University of California, Berkeley examine the effectiveness of symbolic execution for software testing some three decades after its introduction.  While more research is needed to scale this approach to very large programs, they conclude that existing tools have already proved effective in testing and finding errors in a variety of software.
• Researcher William Cheswick sees an “authentication plague upon the land,” and assesses the varied and incomprehensible rules required by “authentication trolls,” many of which date back to the deep past of security concerns.  Citing the changing nature of threats, he poses suggestions to improve the process of securing personal information.
• RSS Laboratories researchers Ari Juels and Alina Oprea address the challenges of migrating enterprise data into the externally-managed public cloud, relying instead on private clouds to retain user trust and visibility.  To extend the data trust perimeter from the enterprise to the public cloud, they have devised cryptographic protocols, and they propose an auditing framework for verification to ensure security and reliability for enterprises.
• Science and Technology writer Neil Savage reports on computational models tackling the complexity of biology, from single-celled microbes to human organs.  He reviews the results of several projects that are using software to wring value from the vast data available to biologists to study how genes interact, how embryos develop, how toxins affect tissues, and why some cells become cancerous.
• George Neville-Neil (aka Kode Vicious) responds to a query about a “best used by” date for software.  Although he likens the idea to freshness marks on milk cartons, he suggests a better analogy for software that involves the erosion and degradation of infrastructure.
• Stanford University’s Steve Cooper and Mehran Sahami reflect on challenges and opportunities for Stanford’s Massive Open Online Courses (MOOCs).  While they believe that MOOCs are here to stay, they do not believe that MOOCs will render teachers obsolete in the foreseeable future.  Their focus, instead, is on the transformative potential of online education if the computing community can find solutions to the challenges at hand.
• Blog @CACM blogger Bertrand Meyer wonders why malicious reviews run rampant in computer science.

For more information on Communications of the ACM, click on cacm.acm.org