SOSP Workshop on Large Scale Distributed Systems and Middleware (LADIS) 2009

October 10-11, 2009

Big Sky Resort
Big Sky, MT

ACM SIGOPS In Cooperation with USENIX
Follow sosp09 on Twitter


Call for Papers Workshop Website Important Dates About the Workshop

LADIS 2009 will bring together researchers and practitioners in the fields of distributed systems and middleware to discuss the challenges of building massive cloud computing infrastructures. By posing research questions in the context of the largest and most-demanding real-world systems, LADIS serves to catalyze dialog between cloud computing engineers and scalable distributed systems researchers, to open the veil of secrecy that has surrounded many cloud computing architectures, and to increase the potential impact of the best research underway in the systems community.

This workshop invites work and promotes the exchange of ideas about:

Particular attention is given to challenges unique to the cloud-computing domain.

The workshop will last for one and a half days, which will include a mix of presentation of accepted papers and as well as keynotes from prominent industry speakers who have been there, made key decisions, and can talk about the architectures of the world's most demanding cloud platforms. LADIS 2008 speakers included Jerry Cuomo (CTO, IBM Web Sphere), James Hamilton (technology guru for Microsoft's Cloud Computing initiative), Franco Travostino and Randy Shoup (CTO and Chief Architect for eBay), and Ben Reed (developer of Yahoo's Zookeeper).

Submission and Logistics

Potential attendees are invited to submit a position or short research paper expressing new ideas, research directions, or relevant opinions. Paper submissions are limited to 5 pages and should include author names and affiliations (i.e., single-blind). The papers should have the same style as SOSP submissions. Papers will be judged on originality, clarity, relevance, and, above all, their likelihood of generating discussion. Accepted papers will be published in the ACM Digital Library.

Workshop Organizers Program Committee

LADIS 2009 Technical Program

Saturday, October 10
1:30 PM - 3:00 PM Keynote #1 and Opening remarks
3:00 PM - 3:30 PM Break
3:30 PM - 4:30 PM Session 1: Programming Models
4:30 PM - 5:30 PM Session 2: Applications and Services
5:30 PM - 6:30 PM Keynote #2
6:30 PM - 8:00 PM Dinner
8:00 PM - 9:30 PM Break-out groups

Sunday, October 11
7:00 AM - 8:30 AM Breakfast
8:30 AM - 9:30 AM Keynote #3
9:30 AM - 10:00 AM Break-out group reports
10:00 AM - 10:30 AM Break
10:30 AM - 12:00 PM Session 3: Storage
12:00 PM - 1:30 PM Lunch
1:30 PM - 3:00 PM Keynote #4
3:00 PM - 3:30 PM Break
3:30 PM - 4:30 PM Session 4: Monitoring and Repair
4:30 PM - 5:15 PM Session 5: Communication
5:15 PM - 5:30 PM Wrap-up

Keynote Speakers       

Keynote #1: Raghu Ramakrishnan, Yahoo!
"Data Management Challenges in the Cloud"

We are in the midst of a computing revolution. As the cost of provisioning hardware and software stacks grows, and the cost of securing and administering these complex systems grows even faster, we're seeing a shift towards computing clouds. For cloud service providers, there is efficiency from amortizing costs and averaging usage peaks. Internet portals like Yahoo! have long offered application services, such as email for individuals and organizations. Companies are now offering services such as storage and compute cycles, enabling higher-level services to be built on top. In this talk, I will discuss Yahoo!'s vision of cloud computing, and describe some of the key initiatives, highlighting the technical challenges involved in designing hosted, multi-tenanted data management systems.

ramakrishnan.img Raghu Ramakrishnan is Chief Scientist for Audience & Cloud Computing, and a Fellow at Yahoo!, where he heads the Community Systems group. He has been Professor of Computer Sciences at the University of Wisconsin-Madison, and was founder and CTO of QUIQ, a company that pioneered question-answering communities, powering Ask Jeeves' AnswerPoint. His research is in the area of database systems, with a focus on data mining, online communities, and web-scale data management. He received the ACM SIGKDD Innovation Award in 2008 and the ACM SIGMOD Contributions Award in 1999. He is a Fellow of the ACM and a Fellow of the IEEE, served as Chair of ACM SIGMOD, is on the Board of Trustees of the VLDB Endowment, and served as editor-in-chief of the Journal of Data Mining and Knowledge Discovery. He received his Ph.D. from the University of Texas at Austin.

Keynote #2: Marvin Theimer, Amazon
"Some Lessons Learned from Running Amazon Web Services"

Running a highly scalable, reliable and cost efficient utility computing platform such as Amazon Web Services involves a number of challenges and opportunities. Designing an architecture that provides resources in an elastic, on-demand manner that is able to withstand fault events such as the loss of entire data centers without impacting a customer's application is just one of the challenges our AWS developers encounter. This talk will focus on core design principles for all AWS services and the lessons learned in acquiring those principles.

theimer.imgMarvin Theimer is a Senior Principal Engineer at Amazon. Marvin received a PhD in Computer Science from Stanford University and has spent most of his career in research, having worked at IBM Almaden, Xerox PARC, and Microsoft Research on topics including distributed operating systems, ubiquitous computing, weakly-consistent replicated systems, peer-to-peer file systems, and global-scale peer-to-peer event notification systems. In 2003 he migrated from research to product development and has worked on web services, grid computing, high-performance computing, eCommerce platforms, and utility computing at Microsoft, Google, and Amazon.

Keynote #3: Jeffrey Dean, Google
"Large-Scale Distributed Systems at Google: Current Systems and Future Directions"

As part of implementing the many products and services offered by Google, we have built a collection of systems and tools that simplify the storing and processing of large-scale data sets, and the construction of heavily-used public services based on these data sets. These systems are intended to work well in Google's computational environment, which consists of large numbers of commodity machines connected by commodity networking hardware. Our systems handle issues like storage reliability and availability in the face of machine failures, and our processing tools make it relatively easy to write robust computations that run reliably and efficiently on thousands of machines. In this talk I'll highlight some of the systems we have built, and discuss some challenges and future directions for new systems.

dean.imgJeff Dean joined Google in 1999 and is currently a Google Fellow in Google's Systems Infrastructure Group. He has co-designed/implemented five generations of Google's crawling, indexing, and query serving systems, and co-designed/implemented major pieces of Google's initial advertising and AdSense for Content systems. He is also a co-designer and co-implementor of Google's distributed computing infrastructure, including the MapReduce and BigTable systems, has worked on system software for statistical machine translation, and implemented on a variety of internal and external developer tools.

Keynote #4: David Nichols, Microsoft
"Life on the Farm: Using SQL for Fun and Profit in Windows Live"

Like other large-scale Internet service companies, Microsoft has an in-house data storage platform supporting transactions, on-line update and a sophisticated query language. It's called SQL Server, and it's deployed on thousands of servers in Windows Live. Using SQL at this scale leads to a number of challenges in areas such as replication, data consistency, backup, reporting, and schema evolution. This talk will look at several of these issues and the tradeoffs we've made in Windows Live to address them.

nichols.imgDavid Nichols is a software developer in the Windows Live group at Microsoft, working on large-scale storage systems. He came to Microsoft by the acquisition of PlaceWare, Inc. where he was a co-founder and principal rchitect. The PlaceWare web conferencing product became Microsoft Office Live Meeting. Before PlaceWare, he was a researcher at Xerox PARC, working on the Tapestry and Jupiter collaboration systems. David has a Ph.D. in Computer Science from Carnegie Mellon University, where he was one of the designers of the Andrew File System.

Session 1: Programming Models       

Cloud-TM: Harnessing the Cloud with Distributed Transactional Memories
Paolo Romano (INESC-ID), Luis Rodrigues (INESC-ID/IST) and Nuno Carvalho(INESC-ID/IST), and Joao Cachopo (INESC-ID/IST)
Storing and Accessing Live Mashup Content in the Cloud
Krzysztof Ostrowski (Cornell University) and Ken Birman (Cornell University)
A Unified Execution Model for Cloud Computing
Eric Van Hensbergen (IBM Research), Noah Evans (Alcatel/Lucent Bell Labs) and Phillip Stanley-Marbell (IBM Research)

Session 2: Applications and Services       

Are Clouds Ready for Large Distributed Applications?
Kunwadee Sripanidkulchai (IBM Research), Sambit Sahu (IBM Research), Yaoping Ruan (IBM Research), Anees Shaikh (IBM Research) and Chitra Dorai (IBM Research)
Cloudifying Source Code Repositories: How Much Does it Cost?
Michael Siegenthaler (Cornell University) and Hakim Weatherspoon (Cornell University)
Cloud9: A Software Testing Service
Liviu Ciortea (EPFL), Cristian Zamfir (EPFL), Stefan Bucur (EPFL), Vitaly Chipounov (EPFL), George Candea (EPFL)

Session 3: Storage       

CRDTs: Consistency without concurrency control
Mihai Leția (ENS Lyon & LIP6), Nuno PreguiƧa (Univ. Nova de Lisboa) and Marc Shapiro (INRIA & LIP6)
Provenance as First Class Cloud Data
Kiran-Kumar Muniswamy-Reddy (Harvard) and Margo Seltzer (Harvard)
Cassandra - A Decentralized Structured Storage System
Avinash Lakshman (Facebook) and Prashant Malik (Facebook)
Towards Decoupling Storage and Computation in Hadoop with SuperDataNodes
George Porter (Sun Labs)

Session 4: Monitoring and Repair       

Toward automatic policy refinement in repair services for large distributed systems
Moises Goldszmidt (Microsoft Research), Mihai Budiu (Microsoft Research), Yue zhang (Microsoft) and Michael Pechuk (Microsoft)
A case for the accountable cloud
Andreas HAeberlen (MPI-SWS)
Learning from the Past for Resolving Dilemmas of Asynchrony
Paul Ezhilchelvan (Newcastle University) and Santosh Shrivastava (Newcastle University)

Session 5: Communication       

Bulletin Board: A Scalable and Robust Eventually Consistent Shared Memory over a Peer-to-Peer Overlay
Vita Bortnikov (IBM Research), Gregory Chockler (IBM Research), Alexey Roytman (IBM Research) and Mike Spreitzer (IBM Research)
Optimizing Information Flow in the Gossip Objects Platform
Ymir Vigfusson (Cornell University), Ken Birman (Cornell University), Qi Huang (Cornell University) and Deepak Parasam Nataraj (Cornell University)