Loading…
This event has ended. View the official site or create your own event → Check it out
This event has ended. Create your own

Cloud Tech is the largest gathering of cloud technologists & engineers in the bay area. Our speakers include the top cloud computing entrepreneurs & experts.

Come join us Saturday, October 6th, from 9am to 6pm at the Computer History Museum in Mountain View, CA for a full 8 hours of learning directly from great minds sharing their secrets!

Register Here

Special thanks to our sponsors who made this all possible. They are: CloudStack, Scalr, VMware, Rackspace, HP, DataStax, AWS, Canonical, Puppet, and General Catalyst.

View analytic

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

Saturday, October 6
 

9:00am

Breakfast & Coffee
Saturday October 6, 2012 9:00am - 9:30am
Hahn Auditorium Computer History Museum

9:00am

Registration
Saturday October 6, 2012 9:00am - 9:30am
Hahn Auditorium Computer History Museum

9:30am

Introduction by Sebastian Stadil

Organizer Sebastian Stadil presents a brief history of the group, and welcomes attendees.


Speakers

Saturday October 6, 2012 9:30am - 9:35am
Hahn Auditorium Computer History Museum

9:35am

Keynote by Jeff Dean: Taming Service Variability, Building Worldwide Systems, and Scaling Deep Learning

In this talk, Jeff describes some techniques and practices for lowering response times in large distributed systems whose components run on shared clusters of machines, where pieces of these systems are subject to interference by other tasks, and where unpredictable latency hiccups are the norm, not the exception. He'll also touch on systems infrastructure for running services that span many geographically distributed data centers. Finally, he'll round out the talk with some highlights of a recent system for scaling deep learning techniques to large amounts of data.


Speakers
avatar for Jeff Dean

Jeff Dean

Google Fellow, Google
Jeff joined Google in 1999 and is currently a Google Fellow in Google's Systems Infrastructure Group. He co-developed the MapReduce computational framework, and is a co-designer and co-implementor of heavily-used distributed storage systems, including BigTable.


Saturday October 6, 2012 9:35am - 10:15am
Hahn Auditorium Computer History Museum

10:15am

Keynote by Andy Bechtolsheim: Networking 2.0

Andy Bechtolsheim, founder of Sun and Arista Networks, first investor in Google, talks about a new generation of networks.


Speakers
avatar for Andreas Bechtolsheim

Andreas Bechtolsheim

Founder, Chief Development Officer and Chairman, Arista Networks
As Chief Development Officer, Andy Bechtolsheim is responsible for the overall product development and technical direction of Arista Networks.\n\nPreviously Andy was a Founder and Chief System Architect at Sun Microsystems, where most recently he was respon


Saturday October 6, 2012 10:15am - 11:00am
Hahn Auditorium Computer History Museum

11:00am

How Percona helps MySQL success in the Cloud?

Learn how Percona helped 100s of customers run MySQL successfully in the cloud.  In this presentation Koa McCullough, DBA at Percona, covers best practices running Percona Server and MySQL in the cloud, looks into setting up backups in the cloud using EBS, Percona Xtrabackup and S3,  using Percona Toolkit to simplify operations, and Percona XtraDB Cluster - the brand new clustering solution designed specially for scaling MySQL application in the cloud, to solve many problems existing in MySQL Replication.

This is a highly technical session with all software being 100% open source.


Speakers
avatar for Koa McCullough

Koa McCullough

Remote DBA, Percona
After he received his B.S. in Computer Science from Montana Tech in\nthe spring of 2006, Koa went to work at RightNow Technologies (RNT).\nAt RNT he has had many titles including System Administrator, MySQL\nDBA, and Cloud Infrastructure Services Developer b


Saturday October 6, 2012 11:00am - 11:45am
Boole Computer History Museum

11:00am

Building Google Cloud Storage
In this talk, Googler Nathan Herring will discuss some of the specific concerns about latency/availabilty tradeoffs in systems choices and network physical layout that were considered when building Google Cloud Storage, and some of the underlying technology used at Google (that's been made public). A little goodness about ops decision-making, and a little insight into the workings at Google to start the day.

Saturday October 6, 2012 11:00am - 11:45am
Hahn Auditorium Computer History Museum

11:00am

Demystifying and Controlling the Performance of Data Center Networks

Although there is tremendous interest in designing improved networks for data centers, very little is known about the network-level traffic characteristics of current data centers.

In this talk, Theophilus Benson presents an empirical study of the network traffic in 10 data centers and show how observations from this study can be used to design more efficient traffic engineering techniques for current data centers.

First, he will describe the flow-level and packet-level transmission properties of the applications within these data centers, and their impact on network utilization, link utilization, congestion, and packet drops.

He will briefly discuss the implications of the observed traffic patterns for data center internal traffic engineering as well as for recently-proposed architectures for data center networks. Next, he will present an overview of MicroTE, a system that adapts to traffic variations by leveraging the short term and partial predictability of the traffic matrix.

He will show that MicroTE performs close to the optimal solution and imposes minimal overhead on the network making it appropriate for current and future data centers.

Finally, he will conclude by discussing the implications of my techniques and observations on application performance.  Towards that end, he will also give a brief overview of ongoing work to provide better visibility into application performance.


Speakers
avatar for Theophilus Benson

Theophilus Benson

Post-Doc, Princeton
Theophilus Benson is passionate about improving the reliability, availability and performance of networks both within local area networks and within the cloud. His work has focused on configuration management, data center networks, and cloud computing.


Saturday October 6, 2012 11:00am - 11:45am
Lovelace Computer History Museum

12:00pm

Services as an Infrastructure: Looking at Enterprise Clouds in a Different Way

Cloud computing has been described in terms of service-style access to "Infrastructure as a Service" (IaaS). In the enterprise, however, it is often better to consider an alternative view in which services themselves are part of theinfrastructure.

In this talk we will discuss "Services as an Infrastructure" and the way in which enterprises can architect production cloud computing environments. Wewill illustrate specific deployment strategies using experiences with Eucalyptus in various enterprise settings and discuss likely future technological developments.


Speakers
avatar for Rich Wolski

Rich Wolski

CTO, Eucalyptus Systems


Saturday October 6, 2012 12:00pm - 12:45pm
Boole Computer History Museum

12:00pm

Actionable Metrics: Enabling Decision-Making in Netflix's Decentralized Environment

In this talk about self-service metrics and alerting, Roy Rapoport describes the approach that Netflix has taken to monitor its production environment. The speaker will walk through the platform they have built and some code samples on how easy it is to setup some of their metrics/alerts. The main point that he will get across is that while each team at Netflix is responsible for their own metrics/alerts, it has been made very easy for them to do so.


Speakers
avatar for Roy Rapoport

Roy Rapoport

Senior DevOps Engineer, Monitoring, Netflix, Inc
Manager, Insights


Saturday October 6, 2012 12:00pm - 12:45pm
Hahn Auditorium Computer History Museum

12:00pm

Packet Transport Mechanisms
 for Data Center Networks

Data centers host diverse applications, mixing workloads that require small predictable latency with others requiring large sustained throughput. In this talk, Mohammad Alizadeh presents measurements from a 6000 server production cluster that reveals impairments with today's state-of-the-art Transmission Control Protocol (TCP) in data center environments. The impairments are rooted in TCP's demands on the limited buffer space available in commodity data center switches. He then discusses Data Center TCP (DCTCP), a new transport mechanism specifically designed for the data center that is now shipping with Windows Server 2012. DCTCP uses a simple modification to the TCP congestion control algorithm that allows it to maintain very low queue occupancy (and latency) in data center switches while simultaneously providing high throughput. Finally, he discusses HULL (for High-bandwidth Ultra-Low Latency), an architecture that builds on DCTCP to deliver baseline fabric latency (only propagation + switching) by nearly eliminating all queueing from the data center fabric.


Speakers
avatar for Mohammad Alizadeh

Mohammad Alizadeh

Ph.D. Candidate, Stanford
Mohammad is a Ph.D. student in the Electrical Engineering department at Stanford University, working with Prof. Balaji Prabhakar. His research interests are broadly in network algorithms and systems. In particular, he is focused on designing high performance packet transport mechanisms for data center networks. Before joining Stanford in 2006, he completed his undergrad at Sharif University of Technology, in Tehran, Iran.


Saturday October 6, 2012 12:00pm - 12:45pm
Lovelace Computer History Museum

1:00pm

Lunch (Thai, see description for menu)

We are pleased to have Buda Thai cater the event, providing us with the following succulent dishes:

  • Spicy Eggplant with chicken
  • Yellow Pumpkin Curry with tofu
  • Pad See Ew with chicken
  • Pineapple Fried Rice
  • Organic Jasmine Rice

Saturday October 6, 2012 1:00pm - 1:45pm
Hahn Auditorium Computer History Museum

2:00pm

Keynote by Raymie Stata: Audit, Patch, and Bootstrap: a rubric for mitigating Murphy's Law

Due to Murphy's law -- more technically: Byzantine failures -- maintaining replicated copies of state is the biggest challenge of building large-scale computer systems. In this talk, Raymie Stata will describe the rubric of "Audit, Patch, and Bootstrap," which he developed at Yahoo to help designers address Byzantine failures in a practical but effective manner.  He'll describe the approach first in the context of more conventional applications.  He'll then describe the approach in the context of systems he has worked on (both inside and outside of Yahoo) that drive IaaS infrastructures like AWS, where Byzantine failures are unfortunately common.


Speakers
avatar for Raymie Stata

Raymie Stata

Founder, VertiCloud
Raymie Stata is CEO of VertiCloud, a startup building a hosted Big Data platform for data-driven applications.  Stata comes to VertiCloud from Yahoo!, Inc., where he was most recently Chief Technical Officer. At Yahoo he worked on algorithmic search, display advertising, and cloud computing.  Stata was an initial champion of Apache Hadoop within Yahoo, and pushed the company to adopt it at larger and larger... Read More →


Saturday October 6, 2012 2:00pm - 2:45pm
Hahn Auditorium Computer History Museum

3:00pm

Reinventing Rackspace — A Lesson in Nimble Cloud Engineering

Building a large-scale OpenStack public cloud is no easy task. Traditional methods of maintenance and upgrading are no longer the norm with the move to the open cloud.  How to do you keep up to date with hundreds of lines of code added daily and a vibrant community of developers growing more and more each day? Come hear Rackspace Developer Advocate Wayne Walls discuss the importance of staying agile but maintaining the focus on the people that matter the most, your customers. He will discuss Rackspace's shift to OpenStack, cloud-on-cloud architecture, and continuous delivery in their relevance to rolling to production faster and more efficiently.


Speakers
avatar for Wayne Walls

Wayne Walls

Director, Cloud Strategy, Rackspace
Wayne Walls is a Sr. Cloud Strategist at Rackspace and accomplished technical leader with over 10 years experience solving problems and developing solutions in startup, enterprise, and academic environments. Often faced with difficult challenges such as IT transformation, technology adoption and consultative selling, Wayne has consistently outperformed expectations by producing highly effective go-to-market plans, strategic implementation, and... Read More →


Saturday October 6, 2012 3:00pm - 3:30pm
Boole Computer History Museum

3:00pm

Escape From Amazon: Tips/Techniques for Reducing AWS Dependencies

As more startups use Amazon Web Services, the following scenario becomes increasingly frequent - the startup is acquired but required by the parent company to move away from AWS and into their own data centers. Given the all encompassing nature of AWS, this is not a trivial task and necessitates careful planning at both the application and systems level. In this presentation, Soam recounts his experience at Delve, a video publishing SaaS platform, with their post acquisition migration to Limelight Networks, a global CDN, during a period of tremendous growth in traffic. In particular, he shares some of the tips/techniques employed during this process to reduce AWS dependence and evolve to a hybrid private/AWS global architecture that has allowed it to compete effectively with digital video leaders like Brightcove and Ooyala.


Speakers
avatar for Soam Acharya

Soam Acharya

Chief Scientist, Limelight Video Platform, Limelight Networks
Soam Acharya is Chief Scientist and Architect of the Limelight Video Platform at Limelight Networks. His current work focuses on video analytics, hybrid cloud architectures and bigdata issues. Prior to acquisition by Limelight, he led Delve Network's initiatives into semantic video, video search, analytics and cloud computing in AWS. | | Previously, Soam worked on web and enterprise search at Inktomi where he was a key member of the GigaDoc... Read More →


Saturday October 6, 2012 3:00pm - 3:45pm
Hahn Auditorium Computer History Museum

3:00pm

Sunny, Rainy and Cloudy Experiences from Clouds for Science

Scientific applications are increasingly investigating cloud models and technologies to manage their computation and data. This talk will cover the results of Magellan, a project funded through the Department
of Energy’s (DOE) Advanced Scientific Computing Research (ASCR) program, evaluated the use of cloud computing for scientific applications. Specifically, Lavanya will cover the outcome of various benchmarking and performance studies, cost analysis and experiences to date and discuss some ongoing efforts to overcome some of the challenges.


Speakers
avatar for Lavanya Ramakrishnan

Lavanya Ramakrishnan

Scientist, Lawrence Berkeley National Lab


Saturday October 6, 2012 3:00pm - 3:45pm
Lovelace Computer History Museum

4:00pm

In the footsteps of a giant — the evolution of CloudStack's architecture

CloudStack got its beginnings in a tiny 10x10 office with its sights set firmly on providing EC2-like functionality to service providers and has since been the platform of choice for more than a hundred public and private production clouds. Along the way, the team faced challenges in understanding and replicating the AWS architecture while winning customers who didn't necessarily want to be AWS. The talk will focus on the architecture choices, dead-ends and successes that CloudStack faced along the way.


Speakers
avatar for Chiradeep Vittal

Chiradeep Vittal

Principal Architect, Citrix Systems
Chiradeep Vittal is Principal Architect in the Cloud Platforms Group at Citrix Systems. He is a maintainer in the Apache CloudStack project where he contributes to networking and storage parts of the Infrastructure-as-a-Service (IAAS) management system. H


Saturday October 6, 2012 4:00pm - 4:45pm
Boole Computer History Museum

4:00pm

A Survey of Petabyte Scale Storage Systems Deployed at Facebook

At Facebook, we use various types of databases and storage system to satisfy the needs of different applications. The solutions built around these data store systems have a common set of requirements: they have to be highly scalable, maintainence costs should be low and they have to perform efficiently. We use a sharded mySQL+memcache solution to support realtime access of tens of petabytes of data and we use TAO to provide consistency of this web-scale database across geographical distances. We use Haystack datastore for storing the 3 billion new photos we host every week. We use Apache Hadoop to mine intelligence from 100 petabytes of clicklogs and combine it with the power of Apache HBase to store all Facebook Messages.

This talk describes the reasons why each of these databases are appropriate for their workloads and the design decisions and tradeoffs that were made while implementing these solutions. We touch upon the consistency, availability and partitioning tolerance of each of these solutions. We touch upon the reasons why some of these systems need ACID semantics and other systems do not. We describe the evolution of our mySQL databases to a pure SSD based deployment.


Speakers
avatar for Dhruba Borthakur

Dhruba Borthakur

Software Engineer, Facebook
Dhruba Borthakur is an engineer in the Database Engineering Team at Facebook. He has been one of the early committers for the Apache Hadoop Distributed File System. He has been associated with Hadoop since its infancy while working for Yahoo. He has been instrumental in scaling Facebook's Hadoop cluster to multiples of petabytes. Dhruba is a contributor to the open source Apache HBase project. Dhruba's current focus is to evaluate alternative... Read More →


Saturday October 6, 2012 4:00pm - 4:45pm
Hahn Auditorium Computer History Museum

5:00pm

Stitching it together: why 10% of infrastructure is 80% of the work

TBD


Speakers
avatar for Charlie Robbins

Charlie Robbins

Director of Engineering, GoDaddy
Charlie is the Gold Director of the Node Foundation, and a Director of Engineering at GoDaddy where he is leading convergence around JavaScript and Node.js across several products through the UX Platform. Charlie was previously the founder and CEO of Nodejitsu (acquired by GoDaddy in 2015). An open source enthusiast and community builder, he is the author of many popular Node libraries, the creator of the EmpireJS and EmpireNode conferences in... Read More →


Saturday October 6, 2012 5:00pm - 5:45pm
Boole Computer History Museum

5:00pm

Running the Largest HDFS Cluster

HDFS is a highly scalable fault-tolerant distributed file system designedfor running on low-cost commodity hardware. HDFS creates multiple replicas of data blocks and distributes them on compute nodes throughout a cluster to enable fast computation on large data sets.


Facebook uses HDFS as a data warehouse, storing Hive tables that collect all Facebook user behaviors from the Facebook's front pages. The warehouse HDFS cluster is consisted of more than thousands of nodes, configured with over 100PB of storage space. Currently the cluster stores hundreds million files, with a growth rate of around hundreds TB of physical space each day. Meanwhile the cluster services a huge load of I/O requests.


The Facebook warehouse HDFS cluster by far is the largest HDFS in the world in term of its capacity. Keeping such a huge file system up andrunning quickly and reliably gives exciting and interesting challenges tothe HDFS team. This talk will present more information on the scale of thecluster, the problems we face, and the solutions we come up to improve theavailability, the scale, and the storage efficiency of the cluster.


Speakers
avatar for Hairong Kuang

Hairong Kuang

Software Engineer, Hadoop committer, Facebook, Apache.org
Hairong Kuang currently leads the development of Hadoop Distributed FileSystem (HDFS) at Facebook. She has been a long time contributor andcommitter to the Apache Hadoop project since she joined Yahoo!. Prior toindustry, she was an Assistant Professor at California State PolytechnicUniversity, Pomona. She received Ph.D. in Computer Science from theUniversity of California at Irvine. Her interests include cloudcomputing, mobile agents, parallel... Read More →


Saturday October 6, 2012 5:00pm - 5:45pm
Hahn Auditorium Computer History Museum