Workshop on Programmability at Cloud Scale 2015


SunriseDistributed computing with the cloud in mind is not easy. Having systems that are reliable and fast is taken for granted these days, but the software required to build them is still hard to use. This workshop will consider reliable, scalable interactive applications for the cloud. There will be talks and discussion of other mature and proposes systems, and of how open source can be utilized to advance and improve tools.

Keynote: Clemens Szyperski


In a world where the focus begins shifting from data being just big to data being also current (volume to velocity), stream processing is gaining importance. Traditional stream processors focus mostly on point-wise transformation, but the notion of complex event processing (CEP), bringing patterns and temporal semantics to the plate, has also been around for almost two decades. Most current systems for stream and complex event processing are presented as programming frameworks, often embracing Java as the core language. Azure Stream Analytics approaches this space differently and fully abstracts operations in a SQL-like language. What easily takes hundreds of lines of Java in many current frameworks like Storm or Spark Streaming collapses to just a few lines of SQL. This talk spans the gamut from why Velocity is a key “V” in the big-data space to how Stream Analytics captures temporal processing declaratively, at language level.


Clemens SzyperskiClemens Szyperski joined Microsoft through its Research division in 1999 and has been in many roles and divisions since. Roles included Principal Architect and Principal Software Development Engineer. Today, he is a Principal Group Software Engineering Manager. He built and runs the team that delivered the Azure Stream Analytics service in April 2015. Before Stream Analytics, Clemens led the effort that yielded the “M” language and underpinning mashup runtime that is used by Power Query for Excel, Power BI, Azure Data Movement Service, and Azure Data Factory. In the space of component technology, his earlier work contributed to the shipping of the Microsoft Extensibility Framework in .NET 4.0. Clemens holds 21 US patents with 10 more pending. He has served on review panels of national funding boards both abroad and in the U.S., including several services for the National Science Foundation. He is on the advisory board for the Masters in Software Engineering course at Seattle University. He is also the U.S. member of the international editorial board of Springer Verlag’s successful Lecture Notes in Business Information Processing, a key outlet for peer-reviewed high-quality proceedings of world-leading academic conferences. Clemens is author or co-author of over one hundred publications that are widely cited (over 6,100 international citations according to ResearchGate). His Jolt-Award winning and best-selling book Component Software is routinely cited as the anchor publication in its field. His award-winning book Software Ecosystem, co-authored with Prof. David Messerschmitt of UC Berkeley, is recognized as the broadest treatment of the economically critical phenomenon of ecosystems formed around software as a product.

Clemens holds a Masters in Electrical Engineering from the Aachen University of Technology (RWTH), Germany, and a PhD in Computer Science from the Swiss Federal Institute of Technology (ETH Zurich), Switzerland. After a postdoc at the International Science Institute (ICSI at UC Berkeley) and co-founding the startup Oberon microsystems, he served as an Associate Professor in the School of Computer Science at the Queensland University of Technology, Brisbane, Australia.



Time Session Speaker
Chair: Phil Bernstein, Microsoft Research
Keynote: Stream Analytics – A brand-new service in Azure and an exercise in high-level platform services
Clemens Szyperski, Microsoft
Lightning Talks
  • Transformations: Software-Defined Distributed Systems, Robbert van Rennesse, Cornell University
  • Orleans Transactions, Phil Bernstein, Microsoft Research
Synapse: A Microservices Architecture for Heterogeneous-Database Cloud Applications Roxana Geambasu, Columbia University
Lightning Talks

  • FollowMyHealth, Joshiah Decker, AllScripts
  • Big Data Debugging, Miryung Kim, UCLA
  • Using Orleans for the Code Hunt Portal, Judith Bishop, Microsoft Research
  • Digital Archeology, Audris Mockis, University of Tennessee
How Orleans Powers Halo Sergey Bykov, Microsoft
Robust Abstractions for Replicated Shared State Sebastian Burckhardt, Microsoft Research
Closing Words Organizers