Monday June 5th, 2007
Before the workshop (8am - noon) there is: Tutorial 1: Serverless Programming (Function as a Service) in room: Charleston 1
Workshop room: Columbia
2:20pm Short break
2:30pm Paper presentations (15min talk + 5 min Q&A)
Ripple: Home Automation for Research Data Management
Pipsqueak: Lean Lambdas with Large Libraries
Leveraging the Serverless Architecture for Securing Linux Containers
3:30pm Afternoon break shared between workshops
4:00pm Paper presentations (15min talk + 5 min Q&A)
Serverless Computing: Design, Implementation, and Performance
Joing us for post-tutorial post-workshop dinner with ICDCS Serverless Workshop participants and serverless developers from Atlanta
(Please signup for meetup so we know how many people to expect and notify you if there are any changes)
Abstract: Serverless architectures let you build and deploy applications and services with utility computing resources that require zero administration. In the past, you would have to provision and scale servers to run your application code, install and operate distributed databases, and build and run custom software to handle API requests. Serverless architectures provide fully-managed services that eliminates these operational complexities. As serverless architectures become more popular, developers need a framework of patterns to help them deploy their workloads without managing servers or operating systems. This talk introduces and describes reusable serverless patterns for web apps, stream processing, batch processing, and automation. For each, we provide a TCO analysis and comparison with its server-based counterpart. We also discuss the considerations and nuances associated with each pattern and share customer experiences. This talk will address how serverless computing is becoming a core component in how companies build and run their applications and services, how serverless computing will continue to evolve, and open research challenges.
Bio: Roger Barga is a General Manager and Director of Development at Amazon Web Services. Roger is also an Affiliate Professor at the University of Washington, where he is a lecturer in the Data Science and Machine Learning programs. Roger holds a PhD in Computer Science, M.Sc. in Computer Science with an emphasis on Machine Learning, and a B.Sc. in Mathematics and Computing Science. He has published over 100 peer-reviewed technical papers, book chapters, along with a book on Machine Learning, and collaborated with over 200 co-authors.
Ryan Chard (Argonne National Laboratory), Kyle Chard (University of Chicago and Argonne National Lab), Jason Alt (National Center for Supercomputing Applications), Dilworth Parkinson (Lawrence Berkeley National Laboratory), Steve Tuecke (University of Chicago and Argonne National Lab), Ian Foster (Argonne National Laboratory & The University of Chicago)
Abstract: Exploding data volumes and acquisition rates, plus ever more complex research processes, place significant strain on research data management processes. It is increasingly common for data to flow through pipelines comprised of dozens of dif- ferent management, organization, and analysis steps distributed across multiple institutions and storage systems. To alleviate the resulting complexity, we propose a home automation approach to managing data throughout its lifecycle, in which users specify via high-level rules the actions that should be performed on data at different times and locations. To this end, we have developed RIPPLE, a responsive storage architecture that allows users to express data management tasks via a rules notation. RIPPLE monitors storage systems for events, evaluates rules, and uses serverless computing techniques to execute actions in response to these events. We evaluate our solution by applying RIPPLE to the data lifecycles of two real-world projects, in astronomy and light source science, and show that it can automate many mundane and cumbersome data management processes.
Edward Oakes (University of Wisconsin-Madison), Leon Yang (University of Wisconsin-Madison), Kevin Houck (University of Wisconsin-Madison), Tyler Harter (Microsoft Gray Systems Lab), Andrea C. Arpaci-Dusseau (University of Wisconsin-Madison), Remzi H. Arpaci-Dusseau (University of Wisconsin-Madison)
Abstract: Microservices are usually fast to deploy because each microservice is small, and thus each can be installed and started quickly. Unfortunately, lean microservices that depend on large libraries will start slowly and harm elasticity. In this paper, we explore the challenges of lean microservices that rely on large libraries in the context of Python packages and the OpenLambda serverless computing platform. We analyze the package types and compressibility of libraries distributed via the Python Package Index and propose PipBench, a new tool for evaluating package support. We also propose Pipsqueak, a package-aware compute platform based on OpenLambda.
Nilton Bila (IBM), Paolo Dettori (IBM), Ali Kanso (IBM), Yuji Watanabe (IBM), Alaa Youssef (IBM)
Abstract: Linux containers present a lightweight solution to package applications into images and instantiate them in isolated environments. Such images may include vulnerabilities that can be exploited at runtime. A vulnerability scanning service can detect these vulnerabilities by periodically scanning the containers and their images for potential threats. When a threat is detected, an event may be generated to (1) quarantine or remove the compromised container(s) and optionally (2) remedy the vulnerability by rebuilding a secure image. We believe that such event-driven process is a great fit to be implemented in a serverless architecture. In this paper we present our design and implementation of a serverless security analytics service based on OpenWhisk and Kubernetes.
Garrett McGrath (University of Notre Dame), Paul R. Brenner (University of Notre Dame)
Abstract: We present the design of a novel performance-oriented serverless computing platform implemented in .NET, deployed in Microsoft Azure, and utilizing Windows containers as function execution environments. Implementation challenges such as function scaling and container discovery, lifecycle, and reuse are discussed in detail. We propose metrics to evaluate the execution performance of serverless platforms and conduct tests on our prototype as well as AWS Lambda, Azure Functions and IBM’s deployment of Apache OpenWhisk. Our measurements show the prototype achieving greater throughput than other platforms at most concurrency levels, and we examine the scaling and instance expiration trends in the implementations. Additionally, we discuss the gaps and limitations in our current design, propose possible solutions, and highlight future research.
Panel chair: Geoffrey C. Fox (Indiana University)
Each panelist will be given a chance to give a five-minute talk about your thoughts, followed by a discussion. Here are some ideas about what you could talk about:
Paul Castro, IBM Research
Vatche Ishakian, Bentley University
Vinod Muthusamy, IBM Research
Aleksander Slominski, IBM Research
Geoffrey C. Fox, Indiana University
Gul Agha, University of Illinois at Urbana-Champaign
Ioana Baldini, IBM Research
Roger Barga, Amazon
Azer Bestavros, Boston University
Flavio Esposito, Saint Louis University
Rodrigo Fonseca, Brown University
Ian Foster, University of Chicago and Argonne National Laboratory
Dennis Gannon, Indiana University & Formerly Microsoft Research
Tyler Harter, Microsoft
Arno Jacobsen, MSRG (Middleware Systems Research Group)
Pietro Michiardi, Eurocom
Peter Pietzuch, Imperial College
Rich Wolski, University of California, Santa Barbara