David Campbell (Microsoft)
Venus & Mars: How to get the traditional systems and cloud computing communities to love each other
October 1, 2013
Many computing communities are colliding in the Cloud Computing shift. In many cases we will converge upon a world incorporating the best of the new and old worlds. One example of this is how Hadoop is evolving to include interactive query capabilities by incorporating techniques from traditional distributed databases.
The convergence of old and new has not been graceful. I've witnessed many examples of the old and new world communities dismissing their mutual value. Each arguing that their approach is better. Thomas Kuhn (The Structure of Scientific Revolution) developed a thesis known as the "theory-dependence of observation". In simple terms it means that what we observe depends on what we know. It is extremely difficult to know what is "better" without a true appreciation for what is different. We look at the same data but observe different things.
As a community we can accelerate our pace of innovation by thoroughly understanding what is truly different in this new world and then melding the best of what we know from the old world with what we've learned in the new.
In this talk I will share many examples of the friction as well as cases where I believe we've broken through as a community. I'll share techniques I've used to help foster shared understanding and broaden our product development efforts. I will close with some thoughts around accelerating our academic and product innovation agenda at large.
Dave Campbell is the CTO of Microsoft's Cloud and Enterprise group. In this role he helps formulate, implement and communicate Microsoft's cloud strategy. Dave has over 25 years of enterprise software experience and has been at Microsoft for 19 years. He has primarily focused on data management challenges and has held a variety of product development roles within Microsoft's SQL Server team during his tenure at Microsoft. Over the last 4 years he has focused on the Azure platform and Microsoft's Big Data strategy.
Dave has been around long enough to have lived through the tectonic computing shift from terminals connected to mainframe and minicomputers to client/server and web scale internet applications. He believes we are only in the first third of the full Cloud computing transition.
Amin Vahdat (Google)
Scale and Programmability in Google's Software Defined Data Center WAN
October 2, 2013
We present the design, implementation, and evaluation of B4, a private WAN connecting Google’s data centers across the planet. B4 has a number of unique characteristics: i) massive bandwidth requirements deployed to a modest number of sites, ii) elastic traffic demand that seeks to maximize average bandwidth, and iii) full control over the edge servers and network, which enables rate limiting and demand measurement at the edge. These characteristics led to a Software De ned Networking architecture using OpenFlow to control relatively simple switches built from merchant silicon. B4’s centralized traffic engineering service drives links to near 100% utilization, while splitting application flows among multiple paths to balance capacity against application priority/demands. We describe experience with three years of B4 production deployment as well as implications of both SDN and WAN deployments on merchant switch silicon.
Amin Vahdat is a Distinguished Engineer at Google where he is the Technical Lead for networking and working on projects in Software Defined Networking and network virtualization. Vahdat's work focuses broadly on computer systems, including distributed systems, networks, and operating systems. He has worked on a range of projects including energy efficiency in data centers, high performance system virtualization, data consistency models for wide-area replication, and data center network architecture. He received a PhD in Computer Science from UC Berkeley and has served on the faculty at Duke University and UC San Diego. Vahdat is an ACM Fellow and a past recipient of the NSF CAREER award and the Alfred P. Sloan Fellowship.
Adrian Cockcroft (Netflix)
Dystopia as a Service
October 3, 2013
We have spent years striving to build perfect apps running on perfect kernels on perfect CPUs connected by perfect networks, but this utopia hasn't really arrived. Instead we live in a dystopian world of buggy apps changing several times a day running on JVMs running on an old version of Linux running on Xen running on something I can't see, that only exists for a few hours, connected by a network of unknown topology and operated by many layers of automation. I will discuss the new challenges and demands of living in this dystopian world of cloud-based services. I will also give an overview of the Netflix open source cloud platform (see netflix.github.com) that we use to create our own island of utopian agility and availability regardless of what is going on underneath.
Adrian is the Director of Architecture for the Cloud Systems team at Netflix, and is leading the Netflix Open Source Software program and the Cloud Prize. Before joining Netflix in 2007, Adrian was a founding member of eBay Research Labs. He spent 16 years at Sun Microsystems, including a stint as Distinguished Engineer and chief architect for Sun's High Performance Technical Computing group. Adrian authored two editions of Sun Performance and Tuning: Java and the Internet; and co-authored two Sun Blueprint books, Resource Management and Capacity Planning for Internet Services.