Musings of a Computer Scientist

August 19, 2015

What is CS, and why study CS? (circa 2000)

Filed under: Uncategorized — Azer Bestavros @ 8:44 am

Here is a blast from the past — a little blurb “About CS” that I put together for our departmental web site some 15 years ago, when I was just starting as chair of computer science at Boston University.


What is Computer Science?

Computer Science is the systematic study of the feasibility, structure, expression, and mechanization of the methodical processes (or algorithms) that underlie the acquisition, representation, processing, storage, communication of, and access to information, whether such information is encoded in bits and bytes in a computer memory or transcribed in genes and protein structures in a human cell.

The fundamental question underlying all of computing is: what computational processes can be efficiently automated and implemented?

To tackle this seemingly simple question, computer scientists work in many complementary areas. They study the very nature of computing to determine which problems are (or are not) computable. They compare various algorithms to determine if they provide a correct and efficient solution to a concrete problem. They design programming languages to enable the specification and expression of such algorithms. They design, evaluate, and build computer systems that can efficiently execute such specifications. And, they apply such algorithms to important application domains.

What Computer Science is Not…

Computer Science is not just about building computers or writing computer programs!

Computer Science is no more about building computers and developing software than astronomy is about building telescopes, biology is about building microscopes, and music is about building musical instruments! Computer science is not about the tools we use to carry out computation. It is about how we use such tools, and what we find out when we do. The solution of many computer science problems may not even require the use of computers — just pencil and paper. As a matter of fact, problems in computer science have been tackled decades before computers were even built. That said, the design and implementation of computing system hardware and software is replete with formidable challenges and fundamental problems that keep computer scientists busy.

Computer Science is about building computers and writing computer programs, and much much more!

Why Computer Science?

In 1943, Thomas J. Watson, Chairman of IBM declared: “I think there is a world market for maybe five computers.” A few billion computers later, there is a temptation to fall into Watson’s embarrassing underestimation of the potential that computing may have on our society. Indeed, in a few decades, “one computer per capita” may sound as outrageous as a “world market of five computers” sounds today. Computer scientists envision a world in which computing is pervasive and seamless.

The golden age of computing (and of computer scientists) has barely begun.

Students choose to major in computer science for a variety of reasons. Many of our students graduate to rewarding computer-related careers in software engineering, system administration and management, research and development in industrial and governmental laboratories. And, since computer technology has transformed almost all disciplines, many of our graduates use their computer science major (and the analytical skills it instills)  to prepare them for a career in other disciplines such as medicine, law, education, physical and life sciences, social sciences, and humanities. Demand for graduates well-versed in computer science is high and is expected to continue to grow as the information age becomes of age!

How Does Computer Science Relate To Scientific Computing?

Computers and software artifacts have become indispensable tools for the pursuit of pretty much every scientific discipline. The use of computers has enabled biologists to comprehend genetics, has enabled astrophysicists to get within femtoseconds of the big bang’s initial conditions, and has enabled geologists to predict earthquakes. It is not surprising, then, for scientists in these disciplines to increasingly rely on a computational methodology (in addition to traditional mathematical or empirical methodologies) to make advances in their respective fields of study. Such scientists are often referred to as computational scientists. So, a computational chemist is a scientist who uses computers to make contribution to chemistry, just as a mathematical physicist uses mathematics to model atomic dynamics, or an empirical biologist uses a microscope to observe cellular behaviors. And, just like all of these scientific disciplines, advances in computer science itself often relies on the use of computers and computational processes. In that sense, among all scientific disciplines, Computer Science is unique. It is the only discipline which fuels its own advancement. Indeed it is a recursive discipline!

How Does Computer Science Relate to Computer Engineering?

The realization of a computing system, subject to various physical and technological constraints, is a challenging undertaking that requires a great deal of knowledge about the functionality and characteristics of the building blocks available at our disposal using today’s technologies (e.g., semiconductor technologies, optical communication technologies, wireless signaling technologies, etc.) Computer engineering concerns itself with current practices in assembling hardware and software components to erect computing engines with the best cost-performance characteristics. In contrast, computer scientists worry about the feasibility and efficiency of solutions to problems in a manner that is less dependent on current technologies. As such, computer scientists work on abstractions that hide details of underlying implementations to enable the construction and comprehension of yet more complex systems. The creative process of developing, implementing, and evaluating computing abstractions is what pushes the frontiers of what computers and computations can do. For example, the pervasive use of the Web in our society is a direct result of our ability to free Internet application developers from the lower-level implementation details of moving bits and bytes over wires from one point to another. Similarly, the tremendous advances in the use of computer animation is a direct result of our ability to free programmers from having to worry about the lower-level details of digital signal processing techniques.

What Does It Take To Be A Successful Computer Scientist?

Computer Science is about problem solving. Thus, the qualities of a good computer scientist include a passion for finding elegant solutions, an ability to use mathematical analysis and logical rigor to evaluate such solutions, creativity in modeling complex problems through the use of abstractions, attention to details and hidden assumptions, an ability to recognize variants of the same problem in different settings, and being able to retarget known efficient solutions to problems in new settings. If you like to solve puzzles, then computer science is for you!

June 3, 2014

Everything you can do I can do better!

Filed under: Uncategorized — Azer Bestavros @ 9:30 am

Essence of the Q&A on a panel that I participated in at Hanscom Air Force Base on Cloud Security:

Google: We take cloud security very seriously. Your information is secure in our cloud. Trust us. 

Me: What if I don't trust you? 

Google: How dare you not trust us? If you don't trust us, who do you trust?

Me: Proofs. Got any?

Google: Anything you can do I can do better; I can do anything better than you.

Me: That's not a proof; that's a song. I can't trust that!

Go figure!

May 2, 2014

Pioneering a “Cloud Mall” for Innovation and Disruption

Filed under: Uncategorized — Azer Bestavros @ 9:42 am

Governor of Massachusetts Deval Patrick, Azer Bestavros, Boston University Rafik B. Hariri Institute for Computing and Computational Science and Engineering, Massachusetts Open Cloud, MOC, Hariri Institute Cloud Computing Initiative, Massachusetts Green High Performance Computing Center

It is truly rewarding to see an idea -- that many have belittled as impractical, unrealistic, and only of academic interest -- finally take on a life of its own, culminating almost seven years of work on the CloudCommons project!

With Massachusetts pledging $3 million in state support, it was wonderful for me to represent the entire BU and MGHPCC team in the proverbial handshake with the Governor of Massachusetts who announced the launch of the Massachusetts Open Cloud (MOC) -- a path-breaking computing cloud that could spur economic growth and technology innovation.

Quoting the story as reported by Bostonia:

The plan calls for hosting the MOC at the MGHPCC data center, where it would tap the computational power of BU and its MOC partners, who have jointly contributed $16 million to MGHPCC, leveraging the $3 million matching grant from the state. Besides the participating universities, MOC partners are tech firms Red Hat, Cisco, EMC, Juniper Networks, SGI, Mellanox, Plexxi, Riverbed, Enterprise DB, Cambridge Computer Services, and DataDirect Networks.

The MOC concept of a cloud marketplace grew out of BU research in 2009. In a recent paper, Bestavros and Krieger argue that closed clouds usually have a single provider, who “alone has access to the operational data.” For this and other reasons, they write, “in the long run, if only a handful of major providers continue to dominate the public cloud marketplace, then any innovation can only be realized through one of them.”

With a cloud designed like the MOC, their paper says, “many stakeholders, rather than just a single provider, participate in implementing and operating the cloud. This creates a multisided marketplace in which participants freely cooperate and compete with each other, and customers can choose among numerous competing services and solutions.”

Another advantage: an open cloud would be more secure than a closed one, Bestavros and Krieger say. It is “the best way to make sure that software is clean,” according to Bestavros, especially as American tech companies complain that federal computer snooping might scare off billions of dollars’ worth of cloud computing customers. With a public, open cloud like the MOC, “the National Security Agency cannot put backdoors in an open-source code, because you can see what the software is doing,” he says.

April 23, 2014

On Social Engineering

Filed under: Uncategorized — Azer Bestavros @ 10:04 am

I never thought that the Boston Globe will quote me on "Social Engineering" -- but that's what they decided to pick out of a long interview for their article “Around Internet, password fatigue setting in Protection becomes a not-so-secret frustration”...

E-mail access in particular is a gateway to other sites because it contains so much personal information, and hackers can exploit a that weak link to work their way into a more lucrative target, such as a bank account, said Azer Bestavros, a professor of computer science at Boston University.

“It’s called ‘social engineering,’ ” he said. “You could be part of a bigger plot. They use you as a step in a bigger scheme. This is why a person who would appear to be totally average could be useful to a hacker.”

March 3, 2014

On Educating Lawmakers and the Public

Filed under: Uncategorized — Azer Bestavros @ 9:52 am

My little contribution to reaching out and "educating" policy makers and the public about the opportunities and challenges of cloud computing, and more importantly, the role that academia plays as an impartial watchdog.

Institute Director, Azer Bestavros, was appointed as board member of the Cloud Computing Caucus Advisory Group. The Cloud Computing Caucus Advisory Group is a non-profit, non-partisan coalition of technology companies and industry groups focused on educating lawmakers and the public about cloud computing as well as other information technology issues. Members of the Board represent some of the largest cloud providers and stakeholders, including Microsoft, Google, Amazon, IBM, EMC, among others.


August 23, 2013

Thank You Paul!

Filed under: Uncategorized — Azer Bestavros @ 10:07 am

Portait of Paul Maritz

A $300,000 gift by Paul Maritz, Chief Executive Officer of Pivotal (an EMC-backed startup), is providing the critical resources needed to launch BU’s Cloud Computing Initiative (CCI), which is incubated at the Hariri Institute for Computing. In a recent visit to the Institute, Paul met with key faculty members involved in CS and CCI projects focusing on novel cloud platforms for high-performance computing, big data, and cyber security. The discussions that ensued helped distill three thrusts required to create “An Advanced Platform for Data Science” at Boston University, which he is generously funding by this gift. Paul’s support is quite important, not only for its financial implications, but also because the imprimatur of his name will open doors and convey recognition of CCI and Institute projects.


August 3, 2013

Supercomputing “for the rest of us!”

Filed under: Uncategorized — Azer Bestavros @ 10:15 am

On July 10, 2013, The Council on Competitiveness, in conjunction with the U.S. House of Representatives Science and National Labs Caucus, held a briefing titled Extreme Computing: Why United States Industry, National Labs, and Academia Need Advanced Computing.

Bestavros speaks at congressional briefing

As one of four speakers, I was invited to participate in this briefing, representing Boston University and academe in general. In my prepared remarks and in follow-up discussions with congressional staff I shared insights on the Massachusetts Green High Performance Center (MGHPCC) and the opportunities that “HPC in the cloud” could offer to society and the importance of securing and hardening our cloud computing infrastructure. The basic tenet of my remarks were that it is about time that HPC and super-computing in general be accessible "to the rest of us" -- as opposed to remaining hostage to the scientific computing community of the last century! Speaking of which, is there really a field of science that is not computational? It's about time we break the mold and make HPC accessible to the myriads of HPC applications that are not being served because they are not serving the "gatekeepers" of scientific computing.

In a thank you note to Boston University, The Council noted that members of the congressional staff and guests were“thrilled to learn of this historic collaboration among the five MGHPCC universities, state government and private industry and its partnership with K-12 public schools and community colleges on new educational and workforce development initiatives.”

July 21, 2013

Social Science Face-lift

Filed under: Uncategorized — Azer Bestavros @ 10:22 pm

A blog post in the New York Times argues that the advent of new research methodologies and the pursuit of new opportunities for exciting interdisciplinary collaboration suggest that social science disciplines are due for an overhaul that resembles the face-lift of many natural science disciplines, which were transformed at least in part by computing.   It's about time! (more...)

June 20, 2013

CS competencies and K12 education: Q&A

Filed under: Uncategorized — Azer Bestavros @ 10:40 am

Massachusetts lawmakers are debating a proposal that would mandate computer science classes in all Massachusetts public schools. The plan would integrate computer skills into the state’s public school curriculum and standardized tests starting as early as the eighth grade. As part of the coverage of this story, I was asked by a local online news outlet to provide my thoughts and comments...

November 18, 2012

Imagine That!

Filed under: Uncategorized — Azer Bestavros @ 11:34 pm

A provocative article by Clay Shirky (writer, consultant and teacher on the social and economic effects of Internet technologies) entitled Napster, Udacity, and the Academy makes parallels between the Music Industry and the Higher Education “industry”.

Quoting from the end of the article:

“In the academy, we lecture other people every day about learning from history. Now its our turn, and the risk is that we’ll be the last to know that the world has changed, because we can’t imagine—really cannot imagine—that story we tell ourselves about ourselves could start to fail. Even when it’s true. Especially when it’s true.”

Imagine that!

[Read the article]


Older Posts »

Powered by WordPress