Home
Videos uploaded by user “Association for Computing Machinery (ACM)”
UIST 2014 Technical Program Preview
 
04:18
A glimpse at the exciting technical program coming up at UIST 2014 in Hawaï, 5-8 October 2014. www.uist.org
CACM Mar. 2018 - A Programmable Programming Language
 
03:22
In the ideal world, software developers would analyze each problem in the language of its domain and then articulate solutions in matching terms. In the real world, however, programmers use a mainstream programming language someone else picked for them. The Racket project seeks to address this problem by utilizing the emerging idea of language-oriented programming. In this video, Matthias Felleisen discusses "A Programmable Programming Language" (cacm.acm.org/magazines/2018/3/225475), a Contributed Article in the March 2018 issue of Communications of the ACM. (Racket is available at http://racket-lang.org/).
Why I Belong to ACM
 
02:26
Bryan Cantrill, Vice President of Engineering at Joyent, Ben Fried, Chief Information Officer at Google, and Theo Schlossnagle, Chief Executive Officer at OmniTI, discuss motivations and benefits of joining the Association for Computing Machinery (ACM). To join ACM: http://www.acm.org/join/professional/PWEBVID More information about ACM: http://www.acm.org
ACM Turing Award 2012
 
01:18
Shafi Goldwasser, Silvio Micali Receive 2012 ACM Turing Award For Advances In Cryptography Shafi Goldwasser and Silvio Micali laid the foundations of modern theoretical cryptography, taking it from a field of heuristics and hopes to a mathematical science with careful definitions and security models, precise specifications of adversarial capabilities, and rigorous reductions from formally defined computational problems. Their results, jointly and with others, established the now-standard definitions of security for the fundamental primitives of encryption and digital signatures, and provided exemplary implementations meeting the stated security objectives. Even more importantly, their work helped to establish the tone and character of modern cryptographic research. Jointly and in collaboration with others, they provided stunning innovations in the form of random functions, interactive proofs, and zero-knowledge protocols, with implications beyond cryptography to theoretical computer science in general. http://amturing.acm.org
Extracting Energy from the Turing Tarpit
 
32:55
Talk by ACM A.M. Turing Laureate Alan C. Kay during the ACM A.M. Turing Centenary Celebration, June, 2012. Abstract: Part of Turing's fame and inspiration came from showing how a simple computer can simulate every other computer, and so "anything is possible". The "Turing Tarpit" is getting caught by "anything is possible but nothing is easy". One way to get caught is to stay close to the underlying machine with our languages so that things seem comprehensible in the small but the code blows up into intractable millions of lines. What if we used "anything is possible" to make very different kinds of computers which require new learning but the code compactly fits the problem and stays small?
CACM Mar. 2016 - Lessons Learned from 30 Years of MINX
 
04:20
Andrew S. Tanenbaum, the author of the MINX operating system, discusses "Lessons Learned from 30 Years of MINIX" (cacm.acm.org/magazines/2016/3/198874), his Contributed Article in the March 2016 CACM.
CACM June 2014 - Leslie Lamport, recipient of the 2013 ACM A.M. Turing Award
 
05:59
ACM's 2013 A.M. Turing Award recipient Leslie Lamport was cited for discovering the field of distributed computing systems that work as intended, making it possible for computers to cooperate, avoid error, and reach consensus. The June 2014 issue of Communications of the ACM details Lamport's innovative advances in an article (cacm.acm.org/news/175166), a Q&A, and an original video highlighting some of his renowned colleagues. In his own voice, he asserts that the best logic for stating things clearly is mathematics, a concept, he notes, that some find controversial. Assessing his body of work, he concludes that he created a path that others have followed to places well beyond his imagination. cacm.acm.org
ACM Author Rights
 
04:07
ACM's expanded Authors' Rights policy offers flexible options that fit computing researchers' individual needs. Authors who publish with ACM can manage their scholarly work through open access pathways to reach the widest possible audience via the ACM Digital Library. Find out how to choose the options that help you take advantage of ACM's reputation as the preferred publisher in the global computing community. Learn more at: http://authors.acm.org/
CACM Feb. 2018 - The Next Phase in the Digital Revolution
 
05:04
Digital Platforms in the computing "cloud" are fundamental features of the digital revolution, entangled with what we term "intelligent tools." An abundance of computing power enabling generation and analysis of data on a scale never before imagined permits the reorganization/transformation of services and manufacturing. How will the increased movement of work to digital platforms provide real and rising incomes with reasonable levels of equality? In this video, John Zysman and Martin Kenney discuss "The Next Phase in the Digital Revolution: Intelligent Tools, Platforms, Growth, Employment," a Contributed Article in the February 2018 issue of Communications of the ACM. Read the full article here: https://cacm.acm.org/magazines/2018/2/224635-the-next-phase-in-the-digital-revolution
Celebrating 50 Years of the ACM A.M. Turing Award and Computing's Greatest Achievements
 
04:59
Since its inauguration in 1966, the ACM A. M. Turing Award has recognized major contributions of lasting importance in computing. Through the years, it has become the most prestigious technical award in the field, often referred to as the “Nobel Prize of computing.” During the next several months, ACM will celebrate 50 years of the Turing Award and the visionaries who have received it. Our aim is to highlight the significant impact of the contributions of the Turing Laureates on computing and society, to look ahead to the future of technology and innovation, and to help inspire the next generation of computer scientists to invent and dream. Our celebration will culminate with a conference on June 23 - 24, 2017 at the Westin St. Francis in San Francisco with lively moderated discussions exploring how computing has evolved and where the field is headed. We hope you can join us there, or via the web—we will be streaming the sessions in real time. For more information please visit, http://www.acm.org/turing-award-50
ACM A.M. Turing Award - Whitfield Diffie and Martin E. Hellman
 
04:03
Whitfield Diffe and Martin Hellman received the 2015 ACM A.M. Turing Award for critical contributions to modern cryptography. The ability for two parties to use encryption to communicate privately over an otherwise insecure channel is fundamental for billions of people around the world. On a daily basis, individuals establish secure online connections with banks, e-commerce sites, email servers and the cloud. Diffie and Hellman's groundbreaking 1976 paper, "New Directions in Cryptography," introduced the ideas of public-key cryptography and digital signatures, which are the foundation for most regularly-used security protocols on the Internet today. The Diffie-Hellman Protocol protects daily Internet communications and trillions of dollars in financial transactions.
Computer Science, Zuckerberg and Video Games
 
02:33
From his February 14, 2013 Google+ Hangout, President Obama discusses the importance of computer science in preparing the nation's future workforce.
CHI2015 Opening Movie
 
01:13
The ACM CHI Conference on Human Factors in Computing Systems is the premier international conference of Human-Computer Interaction. The theme for CHI 2015 in Seoul, Korea, is "Crossings": crossing borders, crossing boundaries, crossing disciplines, crossing people and technology, crossing past and future, crossing physical and digital, crossing art and science, … crossing you and me (http://chi2015.acm.org/).
Bryan Cantrill on why he belongs to ACM
 
02:47
Bryan Cantrill, Vice President of Engineering at Joyent, on ability of the Association for Computing Machinery (ACM) to inspire professional excellence, broaden personal horizons, and bridge the academic/practitioner divide. To join ACM: http://www.acm.org/join/professional/PWEBVID More information about ACM: http://www.acm.org
Program your next server in Go
 
01:02:06
Author: Sameer Ajmani Abstract: Go is a new general-purpose programming language for professionals who build and maintain production systems. Hundreds of companies and thousands of open-source projects are using Go, including Google, DropBox, Docker, Apcera, and SoundCloud. This talk will present Go to the experienced service developer and show how its radically simple approach to software construction can make teams more productive. ACM DL: http://dl.acm.org/citation.cfm?id=2960078 DOI: http://dx.doi.org/10.1145/2959689.2960078
CACM Mar 2015 - Local Laplacian Filters  Edge aware Image Processing with a Laplacian Pyramid HD
 
04:40
Co-author Sylvain Paris discusses "Local Laplacian Filters: Edge-aware Image Processing with a Laplacian Pyramid," the Research Highlights article published in the March 2015 Communications of the ACM (cacm.acm.org/magazines/2015/3/183587).
Inside Websockets
 
39:12
Author: Leah Hanson Abstract: This talk will focus on how WebSockets work -- the details of the protocol and why they are the way they are. Protocol design is about tradeoffs, and if you pick the wrong tradeoff, you may regret it for a very long time. Were going to take a look at the tradeoffs that the WebSockets protocol made and talk about how you can apply the same principles to your own protocols. ACM DL: http://dl.acm.org/citation.cfm?id=2960084 DOI: http://dx.doi.org/10.1145/2959689.2960084
From L3 to seL4 what have we learnt in 20 years of L4 microkernels?
 
33:45
The L4 microkernel has undergone 20 years of use and evolution. It has an active user and developer community, and there are commercial versions which are deployed on a large scale and in safety-critical systems. In this paper we examine the lessons learnt in those 20 years about microkernel design and implementation. We revisit the L4 design papers, and examine the evolution of design and implementation from the original L4 to the latest generation of L4 kernels, especially seL4, which has pushed the L4 model furthest and was the first OS kernel to undergo a complete formal verification of its implementation as well as a sound analysis of worst-case execution times. We demonstrate that while much has changed, the fundamental principles of minimality and high IPC performance remain the main drivers of design and implementation decisions. In the ACM Digital Library: http://dl.acm.org/citation.cfm?id=2522720
John Hennessy and David Patterson 2017 ACM A.M. Turing Award Lecture
 
01:19:38
2017 ACM A.M. Turing Award recipients John Hennessy and David Patterson delivered their Turing Lecture on June 4 at ISCA 2018 in Los Angeles. The lecture took place from 5 - 6 p.m. PDT and was open to the public. Titled “A New Golden Age for Computer Architecture: Domain-Specific Hardware/Software Co-Design, Enhanced Security, Open Instruction Sets, and Agile Chip Development,” the talk will cover recent developments and future directions in computer architecture. Hennessy and Patterson were recognized with the Turing Award for “pioneering a systematic, quantitative approach to the design and evaluation of computer architectures with enduring impact on the microprocessor industry.”
Celebrating 50 Years of Computing's Greatest Achievements
 
02:15
Since its inauguration in 1966, the ACM A. M. Turing Award has recognized major contributions of lasting importance in computing. Through the years, it has become the most prestigious technical award in the field, often referred to as the “Nobel Prize of computing.” During the next several months, ACM will celebrate 50 years of the Turing Award and the visionaries who have received it. Our aim is to highlight the significant impact of the contributions of the Turing Laureates on computing and society, to look ahead to the future of technology and innovation, and to help inspire the next generation of computer scientists to invent and dream. Our celebration will culminate with a conference on June 23 - 24, 2017 at the Westin St. Francis in San Francisco with lively moderated discussions exploring how computing has evolved and where the field is headed. We hope you can join us there, or via the web—we will be streaming the sessions in real time. For more information please visit, http://www.acm.org/turing-award-50
UIST 2015 Technical Program Preview
 
05:12
A glimpse at the exciting technical program coming up at UIST 2015 in Charlotte, 8-11 November 2015. www.uist.org ----------------------------- Music is Big Car Theft by Jason Shaw http://freemusicarchive.org/music/Jason_Shaw/Audionautix_Tech_Urban_Dance/TU-BigCarTheft
Exploring the Reactive Extensions in JavaScript
 
01:00:16
What's does a mouse drag event have in common with an Array of numbers? The answer to this question may surprise you: they are both collections. This key insight holds the key to dramatically simplifying asynchronous programming in JavaScript. In this talk you will learn how you can use the familiar JavaScript Array#extras methods to create surprisingly expressive asynchronous programs. Using just a few functions, you will learn how to do the following: Declaratively build complex events out of simple events (ex. drag n' drop) Coordinate and sequence multiple Ajax requests Reactively update UIs in response to data changes Eliminate memory leaks caused by neglecting to unsubscribe from events Gracefully propagate and handle asynchronous exceptions In this talk we'll be exploring the Reactive Extensions for JavaScript (Rx) library (https://github.com/Reactive-Extensions/RxJS) which allows us to treat events as collections. We'll also contrast RxJS with Promises, CSP and other popular approaches to building asynchronous programs in JavaScript. We'll also dive into the future with RxJS with generators, transducers, and even query transformations. Matthew Podwysocki http://dx.doi.org/10.1145/2742580.2742817
Utilizing the other 80% of your system's performance: Starting with Vectorization
 
01:14:59
Vectorization, as opposed to parallelization, is less utilized as a means of exploiting the full capabilities of a processor. This is a problem since even today this means only ¼ to ½ of the performance of the CPU is used. This is only getting worse in future, especially as accelerators are becoming more prevalent. After an introduction to the basics and history of vectorization the talk will introduce various techniques available for vectorization of compiled code. This talk will focus on gcc and for some details on Linux but the knowledge should be transferable if the features are fully implemented elsewhere. Ulrich Drepper http://dx.doi.org/10.1145/2742580.2742805
CACM June 2015 - Michael Stonebraker 2014 ACM A M Turing Award Recipient
 
04:40
Michael Stonebraker, recipient of the 2014 ACM A.M. Turing Award, bridges the worlds of academia and industry and advocates others do the same. He emphasizes the importance of using academic ideas to build practical applications. Stonebraker discusses his work on INGRES, Edgar Codd's influence, and the state of today's relational database systems in the June 2015 issue of CACM, http://cacm.acm.org/magazines/2015/6.
CACM June 2015 - IllumiRoom Immersive Experiences Beyond the TV Screen
 
05:25
IllumiRoom is a proof-of-concept system that surrounds a television with projected light, bringing video games, and film experiences out of the TV screen and into the real world. It uses 3D scanning and projected light to change the appearance of the room, induce apparent motion, extend the field of view, and enable entirely new gaming experiences. It's entirely self-calibrating and is designed to work in any room. Hrvoje Benko and Brett R. Jones discuss "IllumiRoom: Experiences Beyond the TV Screen," their Research Highlights article in the June 2015 CACM, http://cacm.acm.org/magazines/2015/6/187312-illumiroom/abstract.
The Challenges of Writing a Massive and Complex Go Application
 
01:01:46
Author: Ben Darnell Abstract: We opted for Go when building CockroachDB, a scale-out, relational database, because of its support for libraries, interfaces, and tooling. However, it has come with its own frustrations, often related to performance and synchronization. And as for Cgo, RocksDB, and other critical external libraries, we've had to hunt down or develop creative workarounds to ensure they work well the rest of the toolchain. In this talk, we'll share how we've optimized our memory usage to mitigate issues related to garbage collection and improved our use of channels to avoid deadlocks. We will also share creative techniques to integrate non-Go dependencies into the Go build process. ACM DL: http://dl.acm.org/citation.cfm?id=2960085 DOI: http://dx.doi.org/10.1145/2959689.2960085
CACM August 2016 - Computational Biology in the 21st Century
 
04:01
In the past two decades, biological data sets have become so massive that it has become difficult to analyze them to discover patterns that illuminate underlying biological processes. In this video, Bonnie Berger discusses "Computational Biology in the 21st Century,” a Review Article in the August 2016 Communications of the ACM.(http://cacm.acm.org/magazines/2016/8/205052-computational-biology-in-the-21st-century/fulltext)
CACM Jan 2015 - Distributed Information Processing in Biological and Computational Systems HD
 
05:02
Co-author Saket Navlakha discusses "Distributed Information Processing in Biological and Computational Systems," his Review Article in the January 2015 Communications of the ACM.
There is more consensus in Egalitarian parliaments
 
25:50
This paper describes the design and implementation of Egalitarian Paxos (EPaxos), a new distributed consensus algorithm based on Paxos. EPaxos achieves three goals: (1) optimal commit latency in the wide-area when tolerating one and two failures, under realistic conditions; (2) uniform load balancing across all replicas (thus achieving high throughput); and (3) graceful performance degradation when replicas are slow or crash. Egalitarian Paxos is to our knowledge the first protocol to achieve the previously stated goals efficiently---that is, requiring only a simple majority of replicas to be non-faulty, using a number of messages linear in the number of replicas to choose a command, and committing commands after just one communication round (one round trip) in the common case or after at most two rounds in any case. We prove Egalitarian Paxos's properties theoretically and demonstrate its advantages empirically through an implementation running on Amazon EC2. In the ACM Digital Library: http://dl.acm.org/citation.cfm?id=2517350
CACM July 2015 - Unifying Logic and Probability
 
05:28
Stuart Russell discusses the BLOG (Bayesian logic) language and open-universe probability models, the subject of "Unifying Logic and Probability," his Review Article in the July 2015 Communications of the ACM. Perhaps the most enduring idea from the early days of AI is that of a declarative system reasoning over explicitly represented knowledge with a general inference engine. Such systems require a formal language to describe the real world; and the real world has things in it. For this reason, classical AI adopted first-order logic—the mathematics of objects and relations—as its foundation. http://cacm.acm.org/magazines/2015/7/188745
ACM President Alexander L. Wolf announces the recipients of the 2015 ACM A.M. Turing Award
 
02:15
ACM President Alexander L. Wolf announces recipients of the 2015 ACM A.M. Turing Award.
CACM June 2018 David Patterson and John Hennessy, 2017 ACM A.M.  Turing Award
 
08:13
At a time when "making an impact" can feel like a vague or even overwhelming prospect, it's worth reviewing the accomplishments of two scientists who have done just that: ACM A.M. Turing Award recipients John Hennessy and David Patterson. What began as a simple-sounding insight—that you could improve microprocessor performance by including only instructions that are actually used—blossomed into a paradigm shift as the two honed their ideas in the MIPS (Microprocessor without Interlocked Pipeline Stages) and RISC (Reduced Instruction Set Computer) processors, respectively. A subsequent textbook, Computer Architecture: A Quantitative Approach, introduced generations of students not just to that particular architecture, but to critical principles that continue to guide designers as they balance constraints and strive for maximum efficiency. In this video, Hennessy and Patterson discuss their pioneering work, their partnership, and the future of computer architecture.
Keynote - Systems at Facebook Scale
 
43:37
Behind it's infrastructure serving 1 billion users, the systems abstractions that Facebook uses provide performance and stability for it's services. Ben Maurer will discuss some of the techniques that have made Facebook efficient and reliable ranging from preventing outages with smart queuing to optimizing memory usage for modern processors Ben Maurer http://dx.doi.org/10.1145/2742580.2742644
The History of Rust
 
58:42
Author: Steve Klabnik Abstract: The Rust programming language recently celebrated its one year anniversary since 1.0. While that's not a long time, there were eight years of development before that, which saw radical changes in the language. In this talk, Steve will show off some of Rust's history, with all of the decisions and changes that were made along the way. ACM DL: http://dl.acm.org/citation.cfm?id=2960081 DOI: http://dx.doi.org/10.1145/2959689.2960081
Carver Mead - Semiconductors
 
27:33
ACM97 Speaker: Carver Mead Position: Gordon and Betty Moore Professor of Engineering and Applied Science, California Institute of Technology Talk: Semiconductors Running time: 26 minutes
Keynote - JSON Graph: Reactive REST at Netflix
 
54:58
Every user of a web application wants to believe that all of the data in the cloud is sitting right on their device. Netflix's data platform "JSON Graph" creates this illusion for the web developer. One Model, Available Everywhere. Using an innovative combination of reactive programming techniques and RESTful principles, JSON Graph allows web developers to create a virtual server JSON model for their web application and transparently access it from any cloud-connected device. The Data is the API. Using JSON Graph, Netflix developers retrieve data from the virtual server model the same way they would from an in-memory JSON object. Efficient client/server interactions are ensured by batching concurrent idempotent requests, transparently optimizing requests into point queries, and caching recently-used data. Come learn about the innovative data platform the powers the Netflix UIs, and the new design patterns it enables. Jafar Husain http://dx.doi.org/10.1145/2742580.2742640
CACM Apr. 2016 - Secure Multiparty Computations on Bitcoin
 
04:44
In the April issue of CACM, the Research Highlights article "Secure Multiparty Computations on Bitcoin" (cacm.acm.org/magazines/2016/4/200175), explores using Bitcoin to design fully decentralized protocols that remain secure even if no trusted third party is available. In this video, co-author Stefan Dziembowski discusses this project.
CACM July 2016 - The Rise of Social Bots
 
03:48
A social bot is a computer algorithm that automatically produces content and interacts with humans on social media, trying to emulate and possibly alter their behavior. These bots have become more prevalent on social networking sites in the past few years. In this video, Emilio Ferrara discusses "The Rise of Social Bots" (cacm.acm.org/magazines/2016/7/204021), a Review Article in the July 2016 Communications of the ACM.
Impacto: Simulating Physical Impact by Combining Tactile Stimulation with Electrical Muscle ...
 
00:31
Impacto: Simulating Physical Impact by Combining Tactile Stimulation with Electrical Muscle Stimulation Pedro Lopes, Alexandra Ion, Patrick Baudisch Abstract: We present impacto, a device designed to render the haptic sensation of hitting or being hit in virtual reality. The key idea that allows the small and light impacto device to simulate a strong hit is that it decomposes the stimulus: it renders the tactile aspect of being hit by tapping the skin using a solenoid; it adds impact to the hit by thrusting the user’s arm backwards using electrical muscle stimulation. The device is self-contained, wireless, and small enough for wearable use, thus leaves the user unencumbered and able to walk around freely in a virtual environment. The device is of generic shape, allowing it to also be worn on legs, so as to enhance the experience of kicking, or merged into props, such as a baseball bat. We demonstrate how to assemble multiple impacto units into a simple haptic suit. Participants of our study rated impact simulated using impacto’s combination of solenoid hit and electrical muscle stimulation as more realistic than either technique in isolation. ACM DL: http://dl.acm.org/citation.cfm?id=2807443 DOI: http://dx.doi.org/10.1145/2807442.2807443
"Advances in Deep Neural Networks," at ACM Turing 50 Celebration
 
01:15:46
Deep neural networks can be trained with relatively modest amounts of information and then successfully be applied to large quantities of unstructured data. Deep learning techniques have been applied with great success to areas such as speech recognition, image recognition, natural language processing, drug discovery and toxicology, customer relationship management, recommendation systems, and biomedical informatics. The capabilities of deep neural networks, in some domains, have proven to rival those of human beings. Panelists will explore how deep neural networks are changing our world and our jobs. They will also discuss how things may further change going forward. Moderator: Judea Pearl (2011 Turing Laureate), University of California, Los Angeles Panelists: Michael I. Jordan, University of California, Berkeley Fei-Fei Li, Stanford University Stuart Russell, University of California, Berkeley Ilya Sutskever, OpenAI Raquel Urtasun, University of Toronto
The Power of Abstraction
 
01:18:35
Author: Barbara Liskov Presented as a Keynote during OOPSLA'09, 24th ACM SIGPLAN conference on Object oriented programming systems languages and applications. Citation: For contributions to practical and theoretical foundations of programming language and system design, especially related to data abstraction, fault tolerance, and distributed computing.
On Methodology: Turing Laureates Discuss their Approach to Work
 
09:48
In this video from ACM's Celebration of 50 Years of the A.M. Turing Award, Turing Laureates Andrew Yao, Marvin Minsky, Herbert Simon, Shafi Goldwasser, James Gray, Edmund Clarke and Richard Karp discuss their approach to work and share advice for those who aspire to follow in their footsteps.
The ACM A.M. Turing Award
 
01:53
The A.M. Turing Award, the ACM's most prestigious technical award, is given for major contributions of lasting importance to computing. Recipients are invited to give the annual A.M. Turing Award Lecture. The award is also accompanied by a cash prize of $1,000,000, which in recent years has been underwritten by Google, Inc. This video provides an overview of the origins of the award.
CACM Mar. 2016 - An Interview with Stanford University President John Hennessy
 
04:01
Stanford University President John Hennessy discusses the future of business, technology, and Silicon Valley with UC Berkeley Computer Science Professor David Patterson (cacm.acm.org/magazines/2016/3/198871), featured in the March 2015 Communications of the ACM.
Large-scale cluster management at Google with Borg
 
27:55
Authors: Abhishek Verma, Luis Pedrosa, Madhukar Korupolu, David Oppenheimer, Eric Tune, John Wilkes Abstract: Google's Borg system is a cluster manager that runs hundreds of thousands of jobs, from many thousands of different applications, across a number of clusters each with up to tens of thousands of machines. It achieves high utilization by combining admission control, efficient task-packing, over-commitment, and machine sharing with process-level performance isolation. It supports high-availability applications with runtime features that minimize fault-recovery time, and scheduling policies that reduce the probability of correlated failures. Borg simplifies life for its users by offering a declarative job specification language, name service integration, real-time job monitoring, and tools to analyze and simulate system behavior. We present a summary of the Borg system architecture and features, important design decisions, a quantitative analysis of some of its policy decisions, and a qualitative examination of lessons learned from a decade of operational experience with it. ACM DL: http://dl.acm.org/citation.cfm?id=2741964 DOI: http://dx.doi.org/10.1145/2741948.2741964
Holistic configuration management at Facebook
 
31:58
Authors: Chunqiang Tang, Thawan Kooburat, Pradeep Venkatachalam, Akshay Chander, Zhe Wen, Aravind Narayanan, Patrick Dowell, Robert Karl Abstract: Facebook's web site and mobile apps are very dynamic. Every day, they undergo thousands of online configuration changes, and execute trillions of configuration checks to personalize the product features experienced by hundreds of million of daily active users. For example, configuration changes help manage the rollouts of new product features, perform A/B testing experiments on mobile devices to identify the best echo-canceling parameters for VoIP, rebalance the load across global regions, and deploy the latest machine learning models to improve News Feed ranking. This paper gives a comprehensive description of the use cases, design, implementation, and usage statistics of a suite of tools that manage Facebook's configuration end-to-end, including the frontend products, backend systems, and mobile apps. ACM DL: http://dl.acm.org/citation.cfm?id=2815400.2815401 DOI: http://dx.doi.org/10.1145/2815400.2815401
Past and future of hardware and architecture
 
30:50
Author: David Patterson Abstract: We start by looking back at 50 years of computer architecture, where philosophical debates on instruction sets (RISC vs. CISC, VLIW vs. RISC) and parallel architectures (NUMA vs clusters) were settled with billion dollar investments on both sides. In the second half, we look forward. First, Moore's Law is ending, so the free ride is over software-oblivious increasing performance. Since we've already played the multicore card, the most-likely/only path left is domain-specific processors. The memory system is radically changing too. First, Jim Gray's decade-old prediction is finally true: "Tape is dead; flash is disk; disk is tape." New ways to connect to DRAM and new non-volatile memory technologies promise to make the memory hierarchy even deeper. Finally, and surprisingly, there is now widespread agreement on instruction set architecture, namely Reduced Instruction Set Computers. However, unlike most other fields, despite this harmony has been no open alternative to proprietary offerings from ARM and Intel. RISC-V ("RISC Five") is the proposed free and open champion. It has a small base of classic RISC instructions that run a full open-source software stack; opcodes reserved for tailoring an System-On-a-Chip (SOC) to applications; standard instruction extensions optionally included in an SoC; and it is unrestricted: there is no cost, no paperwork, and anyone can use it. The ability to prototype using ever-more-powerful FPGAs and astonishingly inexpensive custom chips combined with collaboration on open-source software and hardware offers hope of a new golden era for hardware/software systems. ACM DL: http://dl.acm.org/citation.cfm?id=2830903.2830910 DOI: http://dx.doi.org/10.1145/2830903.2830910
ACM Queue Inspirations with Jim Waldo HD
 
04:15
Jim Waldo, Chief Technology Officer at Harvard University, discusses his work on data de-identification, and the question of how to protect user privacy while aggregating accurate models. Download the Queue app here: https://queue.acm.org/app/landing.cfm
CACM Feb. 2018 - Elements of the Theory of Dynamic Networks
 
04:22
We are rapidly approaching the era of dynamicity and of the highly unpredictable. A great variety of modern networked systems are highly dynamic both in space and time. Many traditional approaches and measures for static networks are not adequate for dynamic networks. There is already strong evidence that there is room for the development of a rich theory in this space. In this video, Othon Michail and Paul G. Spirakis discuss "Elements of the Theory of Dynamic Networks," a Review Article in the February 2018 issue of Communications of the ACM. Read the full article here: https://cacm.acm.org/magazines/2018/2/224637-elements-of-the-theory-of-dynamic
CACM September 2018 - A Domain Specific Architecture for Deep Neural Networks
 
05:18
The ending of Moore's Law leaves domain-specific architectures as the future of computing. A trailblazing example is the Google's tensor processing unit (TPU), first deployed in 2015, and that provides services today for more than one billion people. It runs deep neural networks (DNNs) 15 to 30 times faster with 30 to 80 times better energy efficiency than contemporary CPUs and GPUs in similar technologies. In this video, Cliff Young and David Patterson discuss "A Domain-Specific Architecture for Deep Neural Networks," a Contributed Article in the September 2018 issue of Communications of the ACM.