Cloud Native Cost Optimization: Adrian Cockcroft, Battery Ventures (prev. Netflix, eBay, Sun)
For traditional datacenter applications capacity is a fixed upfront cost, so there is little incentive to stop using it once it’s been allocated, and it has to be over-provisioned most of the time so there is enough capacity for peak loads. When traditional application and operating practices are used in cloud deployments there are immediate benefits in speed of deployment, automation, and transparency of costs. The next step is a re-architecture of the application to be cloud native, and significant operating cost reductions can help justify the development work. Cloud native applications are dynamic and use ephemeral resources that are only charged for when they are in use. This talk will discuss best practices for cloud native development, test and production deployment architectures that turn off unused resources and take full advantage of optimizations such as reserved instances and consolidated billing.
Adrian Cockcroft has had a long career working at the leading edge of technology. He’s always been fascinated by what comes next, and he writes and speaks extensively on a range of subjects. At Battery, he advises the firm and its portfolio companies about technology issues and also assists with deal sourcing and due diligence.
Before joining Battery, Adrian helped lead Netflix’s migration to a large scale, highly available public-cloud architecture and the open sourcing of the cloud-native NetflixOSS platform. Prior to that at Netflix he managed a team working on personalization algorithms and service-oriented refactoring.
Adrian was a founding member of eBay Research Labs, developing advanced mobile applications and even building his own homebrew phone, years before iPhone and Android launched. As a distinguished engineer at Sun Microsystems he wrote the best-selling “Sun Performance and Tuning” book and was chief architect for High Performance Technical Computing.
Joint keynote speaker between UCC and the Cloud Control Workshop.
A Few Control Issues in Warehouse-scale Computing: John Wilkes, Google
Google’s warehouse-scale computing infrastructure supports almost all of our external services. The cluster management systems that manage it allocate resources to different applications on our fleet of computers, automate software installations and hardware maintenance, provide monitoring, and do many other things. They rely on control loops at many different granularities. I’ll present an overview of some of these systems and highlight a few of the opportunities and challenges they present, driven by the scale at which we operate, an acute awareness of failures, and the drive to provide ever-better service-levels while curbing complexity.
John Wilkes has been at Google since 2008, where he is working on cluster management and service level agreements for infrastructure services. He is interested in far too many aspects of distributed systems, but a recurring theme has been technologies that allow systems to manage themselves.
He received a PhD in computer science from the University of Cambridge, joined HP Labs in 1982, and was elected an HP Fellow and an ACM Fellow in 2002 for his work on storage system design. Along the way, he’s been PC chair for SOSP, FAST, EuroSys and HotCloud, and serves on the steering committees for SoCC. He’s listed an inventor on 41 US patents, and has an adjunct Professor appointment at Carnegie-Mellon University.
Keynote at the Cloud Control Workshop.
Rack-Scale Computers and the Cloud: Ant Rowstron, Microsoft
In a world where we deploy cloud infrastructure by the rack, why should we be thinking of hardware units that are fractions of a rack? In this talk I will discuss some of the emerging hardware trends and try to argue for the benefits for thinking of designing cloud infrastructure at a rack-scale. I’ll talk about work we have done understanding the impact and opportunities for networking at the rack-scale, and I’ll also describe the a rack-scale cold storage unit that we have built called Pelican. Pelican an example of a rack-scale design, it is not built around the traditional design thinking, but is a design that is right-provisioned at the rack-scale, with just sufficient power, cooling, networking and compute resources to service cold storage workloads. This minimizes costs compared to traditional cloud storage infrastructure, and demonstrates the potential of rack-scale computers.
Ant Rowstron is a Principal Researcher and leads Systems and Networking Group at Microsoft Research. His research interests are broad, covering the spectrum of systems, distributed systems, storage and networking.
He received an MEng degree in Computer Systems and Software Engineering in 1993 from the University of York, UK, and a DPhil degree in Computer Science in 1996 also from the University of York, UK. In 1996 he moved to the Computer Laboratory at Cambridge University, then to the Laboratory for Communications Engineering in the Engineering Department, Cambridge University. In 1999 he moved to Microsoft Research Cambridge.
Keynote at the Cloud Control Workshop.
Infrastructure As Code: Joe Baguley, Chief Technology Officer, EMEA, VMware
As the “cloud” platforms evolve into operating systems for whole datacenters, how will IT skills evolve? What will happen to the industry landscape?
Joe Baguley is VMware’s Chief Technology Officer for EMEA, helping to develop and communicate VMware’s strategy and vision with customers and partners. As part of VMware’s Office of the CTO and its representative in EMEA, Joe assists VMware’s customers in understanding how to use today’s advances in technology to deliver real business impact. Joe is a recognised leader within the European technical community and considered one of the top 50 most influential leaders in UK IT, according to Computer Weekly. He has played a key role in CloudCamp and other events, communicating how cloud technology fits into the broader IT landscape. Joe previously spent ten years at Quest Software where he was CTO of EMEA, shaping its direction and strategy.
In addition, Joe is also on several advisory boards at the European Commission and ETSI and is a founding committee member of the Data Centre Specialist Group at the British Computer Society which helped shape the European Code Of Conduct for Data Centres.
Medical Big Data on the Cloud : the eTRIKS mission: Professor Yike Guo, Imperial College London
Modern medicine is becoming a data-driven science. Improving patient care and developing personalized therapies depends increasingly on an organization’s ability to rapidly and intelligently leverage complex molecular and clinical data from a variety of internal, partner and public sources. In Imperial, we are leading the development of the eTRIKS platform which aims at providing an open source architecture for the analysis of this massive amount of molecular data in the context of personal medicine research. The project mandates the development of a scalable, performant, resilient to failure and flexible solution for medical big data on the cloud. This presentation gives an overview of the eTRIKS project. Then we will focus on the cloud based analytical engine built upon the ICBIG data management system which is newly established Data Science Institute at Imperial College London. Base of this platform is consisting a stack containing Hadoop and Spark and many other frameworks in apache Hadoop ecosystem to provide big data management support. Based on that, we are building an analytical engine to provide high performance medical bioinformatics services for the large scale clinical research for personal medicine.
Professor Yike Guo is the founding director of the multi-disciplinary Data Science Institute at Imperial College London. He has carried out research and development in the field of data intensive analytical computing at Imperial since 1995, when he was the technical director of Imperial’s Parallel Computing Centre. Professor Guo focuses on applying data mining technology to scientific data analysis, in the fields of life science, healthcare, environment science and economy. He has published nearly 200 research papers.
Actors in the Cloud: Supporting Security, Scalability and Availability: Professor Gul Agha, University of Illinois at Urbana-Champaign
Actor programming languages are designed to support parallelism from the ground up, making them suitable for scalable concurrent applications. The actor model facilitates security through isolation of local state and locality properties, and availability through location transparency and migration. Not surprisingly, large-scale applications such as Facebook chat service and Twitter have been written in actor languages. The talk will introduce the actor model and describe issues in the design and implementation of actor languages. A particular focus will be on using actors to address the challenge of coordination and resource management in the cloud.
Gul Agha is Professor of Computer Science and Co-PI in the Assured Cloud Computing Center at the University of Illinois at Urbana-Champaign. His research is focused on programming models and languages for open distributed and embedded computation. Dr. Agha is a Fellow of the Institute for Electrical Engineering and Electronics (IEEE). He served as Editor-in-Chief of IEEE Concurrency: Parallel, Distributed and Mobile Computing (1994-98), and of ACM Computing Surveys (1999-2007). His book on Actors, published by MIT Press, is among the most widely cited works. He has published over 150 research articles and supervised 30 PhD dissertations.
Enabling Cloud Computing at Hyper-Scale: Dr. Kenji Takeda, Microsoft
Cloud computing is pervasive in everyday life, from email and social networks, to smartphones and gaming consoles. It enables an unlimited number of services to connect people, devices, and data in a way that is becoming increasingly seamless. The complexity of achieving this at global scale from both the hardware and software perspectives can be tremendous, raising many research challenges. We describe how Microsoft Research is helping to make Azure a scalable, robust, trustworthy, high performance, hyper-scale cloud platform.
Dr Kenji Takeda is Solutions Architect and Technical Manager for Microsoft Research. His current focus is helping researchers take best advantage of cloud computing, including through big data and data science approaches, including the Azure for Research programme – www.azure4research.com. He has extensive experience in Cloud Computing, High Performance and High Productivity Computing, Data-intensive Science, Scientific Workflows, Scholarly Communication, Engineering and Educational Outreach. He has a passion for developing novel computational approaches to tackle fundamental and applied problems in science and engineering. He was previously Co-Director of the Microsoft Institute for High Performance Computing at the University of Southampton, UK.
Clinical Intelligence & Integration – Tackling Large Data Analytics in Real Time: Oliver Vettel, Andreas Koop, F. Hoffmann-La Roche AG
Understand how Roche Diagnostics is taking clinical intelligence to the next level: the talk includes experiences on tackling ever-changing data models in a validated environment, the inclusion of genomics or proteomics data to stratify patients in a clinical trial and how to take the next step towards real-life data science.
Dr. Andreas Koop is IT program manager within the solution center for research & development at Roche Diagnostics and globally responsible for the clinical intelligence & integration program at Roche. With a background in medical informatics, Andreas has many years of experience in the healthcare industry.
Oliver Vettel is chief technology & information officer at the inhive group, located in Lorsch, Basel and Dubai. As a technology advisor and solution architect to Roche, he leads the execution of the clinical intelligence & integration strategy from a technical perspective.
Headquartered in Basel, Switzerland, Roche is a leader in research-focused healthcare with combined strengths in pharmaceuticals and diagnostics. Roche is the world’s largest biotech company, with truly differentiated medicines in oncology, immunology, infectious diseases, ophthalmology and neuroscience. Roche is also the world leader in in vitro diagnostics and tissue-based cancer diagnostics, and a frontrunner in diabetes management. Roche’s personalised healthcare strategy aims at providing medicines and diagnostics that enable tangible improvements in the health, quality of life and survival of patients. Founded in 1896, Roche has been making important contributions to global health for more than a century. Twenty-four medicines developed by Roche are included in the World Health Organisation Model Lists of Essential Medicines, among them life-saving antibiotics, antimalarials and chemotherapy.
Steve Strutt, CTO for Cloud Computing, IBM UK and Ireland (Thursday 11th in SCCTSA session, from 11:05am)
It is clear cloud, social and mobile innovation is changing the way people interact with and consume information in the urban environment. It is also changing their expectations on the types of services available, speed of access, relevance of information and the pace of change. Together these have major implications for the smarter city and on how the increasing expectations of citizens can be met. This session will explore the evolving world of cloud computing and the new delivery and business models that it enables as they relate to the citizen and smarter city. Though is this future world of interconnected devices and massive datasets sustainable? Along with the opportunities, new thinking will be necessary to address some of the challenges related to the explosion of mobile devices and data.
Steve Strutt is CTO for Cloud Computing, IBM UK and Ireland, working with clients to derive business value from the “as a service” delivery model of cloud computing. He assists clients to innovate in IT service delivery, and has pioneered the adoption of cloud technologies and business models to realise increased client IT productivity, agility and cost reduction. He is passionate about the sustainable use of IT and is an active member of the Intellect Data Centre Council, The Green Grid EMEA technical working group and the BCS.
Steve’s expertise is in engaging with clients at all levels from executives to operations personnel, articulating the value of end-to-end integrated infrastructure and service delivery. His success comes from a deep technical background combined with a strong commercial awareness and an understanding of customers’ needs. His proficiency extends across the full range of IT service delivery, from data centres to virtualized infrastructure and service management, through to application development and delivery, and cloud based delivery models. His passion is working across IT disciplines to deliver solutions, bringing together teams, multiple components or approaches to solve client’s challenges. During his career he has widened his skills from a detailed technical product focus into broader architecture and design. He has experience of working with a wide variety of clients in many industries, demonstrating IBM’s industry leadership while bringing a systems approach to the design of IT solutions.