TUTORIAL T1 – Practical Cloud Deployment and Management with Openstack and Programmable Networking
Date and Period: Monday AM, May 5, 2014
Speakers: Masum Z. Hasan and Horst Dumcke
Room: Gamma 1
Abstract:
Openstack has emerged as a major Cloud Framework (note we use the term "framework" as generalization of Cloud OS, Cloud controller, Cloud Management System, relevant software/API framework) for deploying large scale public and private Clouds. Openstack allows deployment of both computer and storage Clouds with the addition of other value-added services, such as VM image, identity, load-balancing and network related services. Programmable networking or PN (aka SDN) is another area that is changing the landscape of networking in a major way. In a Cloud all the services are on-demand and programmatic. That is, compute and storage resources are requested on-demand and via programmatic interfaces (aka API) with the resources made available to the requester almost immediately or in a short time interval since the request. In general, the CRUD (Create, Read, update, delete) of resources are performed on-demand and programmatically. Cloud requires that network resources are also CRUD on-demand and programmatically. Hence PN or SDN comes in handy in a Cloud environment. In this tutorial we will cover following:
- General Cloud concepts
- Programmable Cloud and networking concepts
- Show how these concepts are realized in the Openstack framework and PN/SDN framework, such as Cisco OnePK and Open source OpenDaylight (withe more focus on Openstack)
- User management and access control framework (Keystone)
- (VM) image management framework (Glance)
- Compute Cloud framework (Nova)
- Elastic block storage framework (Cinder)
- Network services framework (Neutron; this is Openstack version of PN to support PN/SDN of server access network) object storage Cloud framework (Swift).
- Introduction on Cisco OnePK
- Introduction on Open Daylight.
- Deployment of an Openstack controlled Compute Cloud in a multi-node or cluster environment
- This will include neutron controlled network for the Compute Cloud
- Deployment of an Openstack (Swift) controlled Storage Cloud over a pool of distributed storage (distributed on general-purpose computer servers).
- Orchestration of multiple services in Openstack Cloud
- A use case of on-demand QoS for compute/storage Cloud using PN/SDN/OnePK
- Openstack Cloud spanning multiple DCs and MAN/WAN (Seamless Cloud)
- Above will be accompanied by demos on a real Cloud infrastructure and programming examples.
Dr. Masum Z. Hasan is currently a Technology Director at the Cloud CTO office in the Video and Collaboration Group of Cisco Systems, San Jose, USA. He currently leads R&D work in Cloud networking, Network Virtualization and Programmable Networking (aka SDN: Software-defined Networking) and Application centric Networking. His work focus also has been in the area of Grid/High-performance Computing, Healthcare Computing and Networking, Network Analytics, Traffic Engineering, Application and Network resource access control, and Network Management. Masum’s prior job experiences include Principal Investigator at the Bell Labs, USA, Research Scientist at the University of Toronto, Canada and software engineer / computer scientist positions in a number of companies / institutions in Canada and Bangladesh. Masum obtained his MMath and PhD in Computer Science from University of Waterloo, Canada and combined BEng-MEng in Computer Engineering from Odessa National Polytechnic University in Ukraine. Masum has been serving on the organizing and technical program committees of a number of IEEE/IFIP International Conferences and co-editor of a number of journals and magazines. He also serves as the Chair for the IEEE Communications Society's Committee on Cloud Communications and Networking. Visit his personal website for more information.
Horst Dumcke is a senior technical solutions architect for Cisco System. In this capacity, he covers application optimization and data center technologies, working with customers all over Europe, Middle East, Africa, North and South America. Prior to this role of consultant Horst was responsible for developing web-based applications at Cisco. He has 20+ years of experience in the industry, more than 10 of which he spend in software engineering before moving to more customer facing roles. Horst holds a Diploma in Mathematics from University of Heidelberg, Germany.
TUTORIAL T2 – Manageability, Embedded Automation, Network Programming and Autonomic Networking – Contradicting Concepts or Complementary Evolutions?
Date and Period: Monday AM, May 5, 2014
Speakers: Michael Behringer, Patrick Charretour, Joe Clarke, Bruno Klauser, Jason Pfeifer
Room: Gamma 2
Abstract:
While software-defined approaches in many IT domains, including network programming, enjoy a lot of attention currently, evolutions of network-embedded automation are powering an increasing number of innovative operational scenarios. At the same time decades of research, development and investment into basic manageability and operational experience are being leveraged to define control loops which tackle routine tasks autonomously.
Are these independent developments or even competitive concepts? Should ‘pure’ implementations rely on a single approach? Can they be combined to reap all the benefits in parallel? Or might they be an integral part of the bigger picture as networking evolves? Why are software-defined approaches seeing a surge in popularity now despite many of the fundamentals being described since years already? In this tutorial we may not provide the ultimate answers to all of those questions. We will review existing definitions and approaches as well as underlying common concepts across them. Based on experience from real-life deployments we’ll be able to share design considerations and implementation choices as well as emerging common design patterns. As current real-life deployments and early adopters tend towards incremental innovation and selective combinations of approaches, we will explore combinations of the following concepts:
- Application Visibility and Control: aims to optimize application experience in an increasingly dynamic and complex environment where applications across private/hybrid and public cloud, voice/video/collaboration demand proactive management of quality of service (QoS) and the paths of application traffic flows.
- Network Automation: Concepts and lessons learned from the recent steep increase in industry adoption of network-embedded manageability and automation. Review of technologies, current best practices as well as challenges going forward.
- Autonomic Networking: Based on the groundwork laid by research and academia, early validation and adoption of autonomic networking concepts became a reality.
- Network Programming: Seemingly contradicting to some, perceived complementary by others – software defined networking and network programming introduce new degrees of freedom to architect how operational procedures and business applications interact with the network.
Background theory and concept presentation will be combined with practical validation and examples to discuss current practice as well as early adoptions and emerging capabilities – from the perspective of Cisco individual contributors and based on Cisco as well as open source and standardized network technology. Current and future work will be discussed, including open questions, interest areas for research collaboration as well as opportunities for practical validation of upcoming concepts.
Michael H. Behringer works at Cisco Systems as a distinguished engineer, where he focuses on autonomic networking, and network complexity. His previous area was service provider core and security issues, such as MPLS security, NGN security, and Denial-of-Service attack prevention. Prior to joining Cisco in 1998, he worked at the European Internet Service Provider DANTE, based in Cambridge, UK. Here he attained the position of senior network engineer and was responsible for the design and implementation of DANTE's pan-European networks. Michael holds a diploma in computer science from the Technical University of Munich. He is an active member of the IETF, co-chair of the IRTF Network Complexity Research Group, and has published several papers, RFCs, and a book on MPLS VPN security.
Patrick Charretour works at Cisco as a Consulting System Engineer for AVC and Medianet, in the EMEAR Enterprise Networking CTO office, based in Paris, France. He is leading a worldwide adoption team for AVC (application visibility and control). The team is leveraging, testing and adapting AVC Architecture with real customer pain points, having an impact on future developments and also being the official field voice to engineering with SVP executive support. In EMEAR, Patrick is focusing on the development and adoption of AVC (application visibility and control) and Medianet architecture. Key deliverables are working closely with Business Unit, being trusted advisor for Customer and Partners, leveraging and training Account teams to help them increasing business and customer satisfaction, developing specific contents, materials, white-paper, labs and demos. Additionally, Patrick is an official speaker to various internal or external events including Cisco Live. Recently Patrick have also contributed in developing specific consulting approach with customer to transform the way we work with them, focus on Customer Uses Cases, trying to address their Business Transformation Challenges. Prior to join Cisco, Patrick has worked as Sales engineer and Manager in the Collaboration Market, mainly in EMEA and China. Patrick holds a Diploma in Engineering from University of Rennes II and is married with two girls born in 2007 and 2011. Learn and Have fun is the way he like to enjoy his professional life.
As a Global Technical Center engineer, Joe Clarke has contributed to network management products and technologies by finding and fixing bugs, as well as implementing maintenance and troubleshooting components in Cisco Prime, Cisco’s flagship network management suite. Joe helps to support and enhance the embedded automation and programmability technologies, such as the Embedded Event Manager, Tcl, and onePK. He is a top-rated speaker at Cisco’s annual user conference, CiscoLive!, as well as a certified as a CCIE, Sun System Administrator, Sun Network Administrator, Sun Security Administrator, Sun Java certified, and VMware Certified Professional. He has authored numerous technical documents on Cisco network management products and technologies as well as a chapter co-author in the upcoming Springer publication, “Network-Embedded Management and Applications: Understanding Programmable Networking Infrastructure”; and he served as one of the technical editors for the Cisco Press book, "Tcl Scripting for Cisco IOS." He works as the XML lead on the Common Vulnerability and Reporting Framework project, which is a cross-vendor effort to standardize how security advisory data is shared amongst organizations. Joe is co-author of six Cisco patents, including one on leveraging XMPP as a network management and a VPN protocol.
Bruno Klauser works at Cisco as a Consulting Engineer for Network Automation and Programming in the EMEAR Enterprise Networking CTO office, based out of Zurich, Switzerland. Within this field he's focusing on the adoption of programmability, autonomic concepts and embedded network automation. Key Deliverables include white paper and book chapter contributions, proof of concept demonstrations, open-source tutorials, transfer of information seminars and supporting development and architectural groups within Cisco and the industry. Bruno is a member of the IEEE and co-organizer of workshops related to his focus areas. Prior to joining Cisco in 2000, Bruno has worked as Software Engineer, Software Architect and Project Manager in SP network management projects and network management software development teams. He holds a diploma in Software Engineering from University of Applied Sciences Brugg/Windisch, Switzerland as well as a Master of Advanced Studies diploma in Human Computer Interaction Design from University of Basel, Switzerland. Bruno is married and father of a boy born in 2001 and a girl born in 2004.
Jason Pfeifer is a Technical Marketing Engineer in the Network Operating Systems Technology Group for Cisco. He joins the NOSTG technical marketing team with 11 years of experience in the core development of Embedded Event Manager. He has been in the EEM group since its inception, and is a joint patent holder of the product. Jason has created EEM solutions for multiple customers, putting the power of EEM to use solving immediate customer needs.
TUTORIAL T3 – Configuration, Management and Enablement in a Software Defined Platform
Date and Period: Monday PM, May 5, 2014
Speakers: Salman A. Baset, Theo Benson, Sambit Sahu
Room: Gamma 1
Abstract:
Cloud computing has evolved tremendously to become a truly viable virtualized platform for enterprise applications and services. These large-scale virtualized platforms often need complex configurations – both static and dynamic – at compute, storage and network resource levels to meet various demands from application and services. Configuration, enablement and management of such dynamic and virtualized resources and services pose significant research and engineering challenges. Towards these, several interesting areas have evolved such as automated and dynamic configurations, software defined networking, and management across layers for dynamic workloads.
Building upon the following three highly successful tutorials,
1. “Provisioning and management of Enterprise Services in a Cloud”, IM 2011
2. “Cloud Enablement Services: Technologies and Challenges”, NOMS 2012
3. “Leveraging Cloud for Extreme Scale Applications”, IM 2013
This proposed tutorial will focus on the configuration, management and enablement of cloud services in a software defined platform, covering both hands-on technologies as well as recent advancements in these areas.
This tutorial consists of three parts.
1. Configuration and management of software defined platforms
2. Enablement and management of applications on software defined platforms
3. Software defined networks for supporting various workload requirements
The first part will introduce the configuration and management of large-scale software- defined platforms through automation. We shall describe Chef and Puppet, the industry standard tools for automating deployments, and give an illustration of deploying an OpenStack-based cloud using Chef in multiple configurations including software-defined networking. We will also illustrate the need and us of Chef-based solutions for continuous deployment in such platforms (e.g., weekly consuming OpenStack code directly from github). We will also describe how public cloud providers such as Amazon are using Chef tooling to construct “Opsworks Service” for rapid configuration and deployment of services. The second part of the tutorial will focus on the enablement and management of various set of applications and workloads on such software defined platforms. We shall describe the challenges posed by applications in such platforms and recent advancements in supporting application deployment, monitoring and dynamic workload managements. We shall provide hands-on exposure through various tools and services such as Amazon CloudFormation, OpsWorks, Storage Gateway, Redshift etc. The third part will introduce in detail how recent advances in software-defined networks are enabling deployment of new applications and migrations of existing applications into the cloud. Through OpenStack Neutron (the software defined networking component in OpenStack), we will introduce the architecture and features, which allows rapid creation and deployment of SDNs, connecting them to VMs, and deploying applications in these VMs. We will also discuss in detail some of the techniques on debugging SDNs in OpenStack, learned through operational experiences in production deployments.
Topic Contents:
1. Configuration and Management
Salman A. Baset IBM Research Division, Thomas J. Watson Research Center, P.O. Box 218, Yorktown Heights, New York 10598 (electronic mail: sabaset@us.ibm.com). Dr. Salman is working as a Research Staff Member at IBM T. J. Watson Research Center since December 2010. He received a B.S. degree in Computer System Engineer from GIK Institute of Engineering Sciences and Technology, Pakistan in 2001, his M.S. and Ph.D. in Computer Science from Columbia University. His current research at IBM is focused on tracing and fault injection in distributed systems, particularly OpenStack, infrastructure DevOps, over-subscription of physical machine resources, automation of software and operating system updates in the cloud, and cloud SLAs. He has been elected as the Release manager of SPEC OSGcloud benchmarking sub-committee, which is working on standardizing a cloud benchmark. He is also a co-author of cloud benchmark framework report published by SPEC. Previously, he was involved in the Internet Engineering Task Force (IETF) and is a coauthor of RELOAD protocol for building peer-to-peer communication systems. He is a recipient of Young Scholars Award by Marconi Society in 2008 and a best paper award at IPTCOMM in 2010. Currently, he is the chair of Distributed and Fault Tolerant Computing (DFTC) Professional Interest Community (PIC) at IBM.
Theo Benson Department of Computer Science, Duke University, D342 LSRC Building, P.O. Box 90129, Durham, North Carolina, 27707 (electronic mail: tbenson@cs.duke.edu). Theophilus Benson is passionate about eliminating the complexity of managing networks and tackling performance oriented problems both within data centers and clouds. His research focuses on Software Defined Networking, infrastructures for big data, and abstractions for managing various workloads in the cloud. This work has earned him IBM fellowships, a best paper award at IMC 2010, and, more recently, his cloud computing platform was acquired by a large cloud provider. He is an Assistant Professor at Duke University. He received his PhD from the University of Wisconsin – Madison in 2012 after which he spent a year at Princeton University as a Post-Doc with Jennifer Rexford. Prior to that, he received his B.S. at Tufts University and worked as a software engineer at an MIT based startup in Waltham, MA.
Sambit Sahu: IBM Research Division, Thomas J. Watson Research Center, P.O. Box 218, Yorktown Heights, New York 10598 (electronic mail: sambits@us.ibm.com). Dr. Sahu has been a research staff member at IBM since 2000. Dr. Sahu received his Ph.D. degree in Computer Science from University of Massachusetts at Amherst. After joining IBM Research, Dr. Sahu has focused on systems and network services management, content distribution network and data center networking and more recently on cloud computing and smarter planet solutions. He has more than 50 technical papers and 60 patents filed in these areas. He is the founding chair of Usenix/ACM HotClpoud workshop which has been well received in the Cloud/Systems community. Dr. Sahu has a best paper award at ACM Internet Measurement Conference 2006. Dr. Sahu’s recent research has been on designing cloud platform and solutions for deploying smarter city services at city-wide scale. He is a master inventor at IBM Research and has Outstanding Achievement Award and Research Division Awards for his pioneering work in these areas.
TUTORIAL T4 – Wearable Computers: a Holistic Design Approach
Date and Period: Monday PM, May 5, 2014
Speakers: Roozbeh Jafari, Hassan Ghasemzadeh
Room: Gamma 2
Abstract:
Wearable computers bring to fruition many opportunities to continuously monitor human body with sensors placed on body. They provide new avenues to continuously monitor individuals, whether it is intended to detect an early onset of a disease or to assess the effectiveness of the treatment. In the past few years, the community has observed a large number of applications that have been developed using wearable computers. Yet, not many have been deployed in a large scale. There are still several challenges that need to be addressed before realizing the ubiquitous use of wearable computers.
In this tutorial, we present a review of wearable computing systems and their applications. We highlight several components of wearable computers including signal processing, software, hardware architectures, sensors and actuators. We will in particular review signal processing techniques suitable for time-series data acquired from wearable sensors namely dynamic time warping and hidden Markov models. We provide a review for existing software platforms for wearable computers along with the current state-of-the-art hardware architectures. We emphasize the need for holistic approaches optimizing and enhancing the performance of wearable computers (e.g., reducing their form). We will review several cross-layered techniques aimed at creating hardware accelerators for wearable computers. We will conclude the talk by highlighting opportunities and future directions in wearable computer design.
Roozbeh Jafari (www.essp.utdallas.edu) is an associate professor at UT-Dallas. He received his PhD in Computer Science (UCLA) and completed a postdoctoral fellowship at UC-Berkeley. His research interest lies in the area of wearable computer design and signal processing. His research has been funded by the NSF, NIH, DoD - TATRC, AFRL, AFOSR, DARPA, SRC and industry (Texas Instruments, Tektronix, Samsung, Telecom Italia & Texas Health Resources). He has published over 100 papers in refereed journals and conferences. He has served as the general and technical program committee chairs for several flagship conferences in the area of Wireless Health and Wearable Computers including the ACM Wireless Health 2012 and 2013, International Conference on Body Sensor Networks 2011 and International Conference on Body Area Networks 2011. He is an associate editor for the IEEE Sensors Journal and the IEEE Internet of Things Journal. He is the recipient of the NSF CAREER award (2012) and the RTAS 2011 best paper award.
Hassan Ghasemzadeh is currently an Assistant Professor of Electrical Engineering and Computer Science at Washington State University (WSU). His research interests lie in different aspects of Embedded System Design including sustainable and green computing, low-power architectures, reconfigurable computing, and system-level optimization. The focus of his current work is on collaborative signal and information processing, power optimization, data analytics, and algorithm design for networked embedded systems with a primary emphasis on applications in healthcare and wellness. His research spans the areas of embedded systems, computer architecture, signal processing, and machine learning, where he has published more than 80 technical papers on the subject. Prior to joining WSU, he was a Research Manager at the UCLA Wireless Health Institute and an Adjunct Professor of Biomedical Informatics at San Diego State University. He was on the faculty of Azad University from 2003-2006 where he served as Chair of Computer Engineering Department at Damavand branch, Tehran, Iran. Dr. Ghasemzadeh received his Ph.D. degree in Computer Engineering from the University of Texas at Dallas in 2010, and spent the academic year 2010-2011 as a Postdoctoral Fellow at the West Health Institute. He received his M.S. degree in Computer Engineering from University of Tehran, Tehran, Iran, in 2001 and his B.S. degree in Computer Engineering from Sharif University of Technology, Tehran, Iran in 1998.
Date and Period: Monday AM, May 5, 2014
Speakers: Masum Z. Hasan and Horst Dumcke
Room: Gamma 1
Abstract:
Openstack has emerged as a major Cloud Framework (note we use the term "framework" as generalization of Cloud OS, Cloud controller, Cloud Management System, relevant software/API framework) for deploying large scale public and private Clouds. Openstack allows deployment of both computer and storage Clouds with the addition of other value-added services, such as VM image, identity, load-balancing and network related services. Programmable networking or PN (aka SDN) is another area that is changing the landscape of networking in a major way. In a Cloud all the services are on-demand and programmatic. That is, compute and storage resources are requested on-demand and via programmatic interfaces (aka API) with the resources made available to the requester almost immediately or in a short time interval since the request. In general, the CRUD (Create, Read, update, delete) of resources are performed on-demand and programmatically. Cloud requires that network resources are also CRUD on-demand and programmatically. Hence PN or SDN comes in handy in a Cloud environment. In this tutorial we will cover following:
- General Cloud concepts
- Programmable Cloud and networking concepts
- Show how these concepts are realized in the Openstack framework and PN/SDN framework, such as Cisco OnePK and Open source OpenDaylight (withe more focus on Openstack)
- User management and access control framework (Keystone)
- (VM) image management framework (Glance)
- Compute Cloud framework (Nova)
- Elastic block storage framework (Cinder)
- Network services framework (Neutron; this is Openstack version of PN to support PN/SDN of server access network) object storage Cloud framework (Swift).
- Introduction on Cisco OnePK
- Introduction on Open Daylight.
- Deployment of an Openstack controlled Compute Cloud in a multi-node or cluster environment
- This will include neutron controlled network for the Compute Cloud
- Deployment of an Openstack (Swift) controlled Storage Cloud over a pool of distributed storage (distributed on general-purpose computer servers).
- Orchestration of multiple services in Openstack Cloud
- A use case of on-demand QoS for compute/storage Cloud using PN/SDN/OnePK
- Openstack Cloud spanning multiple DCs and MAN/WAN (Seamless Cloud)
- Above will be accompanied by demos on a real Cloud infrastructure and programming examples.
Dr. Masum Z. Hasan is currently a Technology Director at the Cloud CTO office in the Video and Collaboration Group of Cisco Systems, San Jose, USA. He currently leads R&D work in Cloud networking, Network Virtualization and Programmable Networking (aka SDN: Software-defined Networking) and Application centric Networking. His work focus also has been in the area of Grid/High-performance Computing, Healthcare Computing and Networking, Network Analytics, Traffic Engineering, Application and Network resource access control, and Network Management. Masum’s prior job experiences include Principal Investigator at the Bell Labs, USA, Research Scientist at the University of Toronto, Canada and software engineer / computer scientist positions in a number of companies / institutions in Canada and Bangladesh. Masum obtained his MMath and PhD in Computer Science from University of Waterloo, Canada and combined BEng-MEng in Computer Engineering from Odessa National Polytechnic University in Ukraine. Masum has been serving on the organizing and technical program committees of a number of IEEE/IFIP International Conferences and co-editor of a number of journals and magazines. He also serves as the Chair for the IEEE Communications Society's Committee on Cloud Communications and Networking. Visit his personal website for more information.Horst Dumcke is a senior technical solutions architect for Cisco System. In this capacity, he covers application optimization and data center technologies, working with customers all over Europe, Middle East, Africa, North and South America. Prior to this role of consultant Horst was responsible for developing web-based applications at Cisco. He has 20+ years of experience in the industry, more than 10 of which he spend in software engineering before moving to more customer facing roles. Horst holds a Diploma in Mathematics from University of Heidelberg, Germany.TUTORIAL T2 – Manageability, Embedded Automation, Network Programming and Autonomic Networking – Contradicting Concepts or Complementary Evolutions?
Date and Period: Monday AM, May 5, 2014
Speakers: Michael Behringer, Patrick Charretour, Joe Clarke, Bruno Klauser, Jason Pfeifer
Room: Gamma 2
Abstract:
While software-defined approaches in many IT domains, including network programming, enjoy a lot of attention currently, evolutions of network-embedded automation are powering an increasing number of innovative operational scenarios. At the same time decades of research, development and investment into basic manageability and operational experience are being leveraged to define control loops which tackle routine tasks autonomously.
Are these independent developments or even competitive concepts? Should ‘pure’ implementations rely on a single approach? Can they be combined to reap all the benefits in parallel? Or might they be an integral part of the bigger picture as networking evolves? Why are software-defined approaches seeing a surge in popularity now despite many of the fundamentals being described since years already? In this tutorial we may not provide the ultimate answers to all of those questions. We will review existing definitions and approaches as well as underlying common concepts across them. Based on experience from real-life deployments we’ll be able to share design considerations and implementation choices as well as emerging common design patterns. As current real-life deployments and early adopters tend towards incremental innovation and selective combinations of approaches, we will explore combinations of the following concepts:
- Application Visibility and Control: aims to optimize application experience in an increasingly dynamic and complex environment where applications across private/hybrid and public cloud, voice/video/collaboration demand proactive management of quality of service (QoS) and the paths of application traffic flows.
- Network Automation: Concepts and lessons learned from the recent steep increase in industry adoption of network-embedded manageability and automation. Review of technologies, current best practices as well as challenges going forward.
- Autonomic Networking: Based on the groundwork laid by research and academia, early validation and adoption of autonomic networking concepts became a reality.
- Network Programming: Seemingly contradicting to some, perceived complementary by others – software defined networking and network programming introduce new degrees of freedom to architect how operational procedures and business applications interact with the network.
Background theory and concept presentation will be combined with practical validation and examples to discuss current practice as well as early adoptions and emerging capabilities – from the perspective of Cisco individual contributors and based on Cisco as well as open source and standardized network technology. Current and future work will be discussed, including open questions, interest areas for research collaboration as well as opportunities for practical validation of upcoming concepts.
Michael H. Behringer works at Cisco Systems as a distinguished engineer, where he focuses on autonomic networking, and network complexity. His previous area was service provider core and security issues, such as MPLS security, NGN security, and Denial-of-Service attack prevention. Prior to joining Cisco in 1998, he worked at the European Internet Service Provider DANTE, based in Cambridge, UK. Here he attained the position of senior network engineer and was responsible for the design and implementation of DANTE's pan-European networks. Michael holds a diploma in computer science from the Technical University of Munich. He is an active member of the IETF, co-chair of the IRTF Network Complexity Research Group, and has published several papers, RFCs, and a book on MPLS VPN security. Patrick Charretour works at Cisco as a Consulting System Engineer for AVC and Medianet, in the EMEAR Enterprise Networking CTO office, based in Paris, France. He is leading a worldwide adoption team for AVC (application visibility and control). The team is leveraging, testing and adapting AVC Architecture with real customer pain points, having an impact on future developments and also being the official field voice to engineering with SVP executive support. In EMEAR, Patrick is focusing on the development and adoption of AVC (application visibility and control) and Medianet architecture. Key deliverables are working closely with Business Unit, being trusted advisor for Customer and Partners, leveraging and training Account teams to help them increasing business and customer satisfaction, developing specific contents, materials, white-paper, labs and demos. Additionally, Patrick is an official speaker to various internal or external events including Cisco Live. Recently Patrick have also contributed in developing specific consulting approach with customer to transform the way we work with them, focus on Customer Uses Cases, trying to address their Business Transformation Challenges. Prior to join Cisco, Patrick has worked as Sales engineer and Manager in the Collaboration Market, mainly in EMEA and China. Patrick holds a Diploma in Engineering from University of Rennes II and is married with two girls born in 2007 and 2011. Learn and Have fun is the way he like to enjoy his professional life. As a Global Technical Center engineer, Joe Clarke has contributed to network management products and technologies by finding and fixing bugs, as well as implementing maintenance and troubleshooting components in Cisco Prime, Cisco’s flagship network management suite. Joe helps to support and enhance the embedded automation and programmability technologies, such as the Embedded Event Manager, Tcl, and onePK. He is a top-rated speaker at Cisco’s annual user conference, CiscoLive!, as well as a certified as a CCIE, Sun System Administrator, Sun Network Administrator, Sun Security Administrator, Sun Java certified, and VMware Certified Professional. He has authored numerous technical documents on Cisco network management products and technologies as well as a chapter co-author in the upcoming Springer publication, “Network-Embedded Management and Applications: Understanding Programmable Networking Infrastructure”; and he served as one of the technical editors for the Cisco Press book, "Tcl Scripting for Cisco IOS." He works as the XML lead on the Common Vulnerability and Reporting Framework project, which is a cross-vendor effort to standardize how security advisory data is shared amongst organizations. Joe is co-author of six Cisco patents, including one on leveraging XMPP as a network management and a VPN protocol. Bruno Klauser works at Cisco as a Consulting Engineer for Network Automation and Programming in the EMEAR Enterprise Networking CTO office, based out of Zurich, Switzerland. Within this field he's focusing on the adoption of programmability, autonomic concepts and embedded network automation. Key Deliverables include white paper and book chapter contributions, proof of concept demonstrations, open-source tutorials, transfer of information seminars and supporting development and architectural groups within Cisco and the industry. Bruno is a member of the IEEE and co-organizer of workshops related to his focus areas. Prior to joining Cisco in 2000, Bruno has worked as Software Engineer, Software Architect and Project Manager in SP network management projects and network management software development teams. He holds a diploma in Software Engineering from University of Applied Sciences Brugg/Windisch, Switzerland as well as a Master of Advanced Studies diploma in Human Computer Interaction Design from University of Basel, Switzerland. Bruno is married and father of a boy born in 2001 and a girl born in 2004. Jason Pfeifer is a Technical Marketing Engineer in the Network Operating Systems Technology Group for Cisco. He joins the NOSTG technical marketing team with 11 years of experience in the core development of Embedded Event Manager. He has been in the EEM group since its inception, and is a joint patent holder of the product. Jason has created EEM solutions for multiple customers, putting the power of EEM to use solving immediate customer needs. TUTORIAL T3 – Configuration, Management and Enablement in a Software Defined Platform
Date and Period: Monday PM, May 5, 2014
Speakers: Salman A. Baset, Theo Benson, Sambit Sahu
Room: Gamma 1
Abstract:
Cloud computing has evolved tremendously to become a truly viable virtualized platform for enterprise applications and services. These large-scale virtualized platforms often need complex configurations – both static and dynamic – at compute, storage and network resource levels to meet various demands from application and services. Configuration, enablement and management of such dynamic and virtualized resources and services pose significant research and engineering challenges. Towards these, several interesting areas have evolved such as automated and dynamic configurations, software defined networking, and management across layers for dynamic workloads.
Building upon the following three highly successful tutorials,
1. “Provisioning and management of Enterprise Services in a Cloud”, IM 2011
2. “Cloud Enablement Services: Technologies and Challenges”, NOMS 2012
3. “Leveraging Cloud for Extreme Scale Applications”, IM 2013
This proposed tutorial will focus on the configuration, management and enablement of cloud services in a software defined platform, covering both hands-on technologies as well as recent advancements in these areas.
This tutorial consists of three parts.
1. Configuration and management of software defined platforms
2. Enablement and management of applications on software defined platforms
3. Software defined networks for supporting various workload requirements
The first part will introduce the configuration and management of large-scale software- defined platforms through automation. We shall describe Chef and Puppet, the industry standard tools for automating deployments, and give an illustration of deploying an OpenStack-based cloud using Chef in multiple configurations including software-defined networking. We will also illustrate the need and us of Chef-based solutions for continuous deployment in such platforms (e.g., weekly consuming OpenStack code directly from github). We will also describe how public cloud providers such as Amazon are using Chef tooling to construct “Opsworks Service” for rapid configuration and deployment of services. The second part of the tutorial will focus on the enablement and management of various set of applications and workloads on such software defined platforms. We shall describe the challenges posed by applications in such platforms and recent advancements in supporting application deployment, monitoring and dynamic workload managements. We shall provide hands-on exposure through various tools and services such as Amazon CloudFormation, OpsWorks, Storage Gateway, Redshift etc. The third part will introduce in detail how recent advances in software-defined networks are enabling deployment of new applications and migrations of existing applications into the cloud. Through OpenStack Neutron (the software defined networking component in OpenStack), we will introduce the architecture and features, which allows rapid creation and deployment of SDNs, connecting them to VMs, and deploying applications in these VMs. We will also discuss in detail some of the techniques on debugging SDNs in OpenStack, learned through operational experiences in production deployments.
Topic Contents:
1. Configuration and Management
- Overview of Chef and Puppet
- Chef-based automated deployment and upgrade – OpenStack example
- Opsworks for Amazon
- Challenges in workload deployment and management
- Recent progresses in automation of workload deployment
- Hands-on exposure to various automation concepts through AWS CloudFormation, OpsWorks, Storage Gateway, Redshift
- SDN overview in OpenStack and challenges
- Diagnosing problems in OpenStack-based SDNs
Salman A. Baset IBM Research Division, Thomas J. Watson Research Center, P.O. Box 218, Yorktown Heights, New York 10598 (electronic mail: sabaset@us.ibm.com). Dr. Salman is working as a Research Staff Member at IBM T. J. Watson Research Center since December 2010. He received a B.S. degree in Computer System Engineer from GIK Institute of Engineering Sciences and Technology, Pakistan in 2001, his M.S. and Ph.D. in Computer Science from Columbia University. His current research at IBM is focused on tracing and fault injection in distributed systems, particularly OpenStack, infrastructure DevOps, over-subscription of physical machine resources, automation of software and operating system updates in the cloud, and cloud SLAs. He has been elected as the Release manager of SPEC OSGcloud benchmarking sub-committee, which is working on standardizing a cloud benchmark. He is also a co-author of cloud benchmark framework report published by SPEC. Previously, he was involved in the Internet Engineering Task Force (IETF) and is a coauthor of RELOAD protocol for building peer-to-peer communication systems. He is a recipient of Young Scholars Award by Marconi Society in 2008 and a best paper award at IPTCOMM in 2010. Currently, he is the chair of Distributed and Fault Tolerant Computing (DFTC) Professional Interest Community (PIC) at IBM. Theo Benson Department of Computer Science, Duke University, D342 LSRC Building, P.O. Box 90129, Durham, North Carolina, 27707 (electronic mail: tbenson@cs.duke.edu). Theophilus Benson is passionate about eliminating the complexity of managing networks and tackling performance oriented problems both within data centers and clouds. His research focuses on Software Defined Networking, infrastructures for big data, and abstractions for managing various workloads in the cloud. This work has earned him IBM fellowships, a best paper award at IMC 2010, and, more recently, his cloud computing platform was acquired by a large cloud provider. He is an Assistant Professor at Duke University. He received his PhD from the University of Wisconsin – Madison in 2012 after which he spent a year at Princeton University as a Post-Doc with Jennifer Rexford. Prior to that, he received his B.S. at Tufts University and worked as a software engineer at an MIT based startup in Waltham, MA.Sambit Sahu: IBM Research Division, Thomas J. Watson Research Center, P.O. Box 218, Yorktown Heights, New York 10598 (electronic mail: sambits@us.ibm.com). Dr. Sahu has been a research staff member at IBM since 2000. Dr. Sahu received his Ph.D. degree in Computer Science from University of Massachusetts at Amherst. After joining IBM Research, Dr. Sahu has focused on systems and network services management, content distribution network and data center networking and more recently on cloud computing and smarter planet solutions. He has more than 50 technical papers and 60 patents filed in these areas. He is the founding chair of Usenix/ACM HotClpoud workshop which has been well received in the Cloud/Systems community. Dr. Sahu has a best paper award at ACM Internet Measurement Conference 2006. Dr. Sahu’s recent research has been on designing cloud platform and solutions for deploying smarter city services at city-wide scale. He is a master inventor at IBM Research and has Outstanding Achievement Award and Research Division Awards for his pioneering work in these areas. TUTORIAL T4 – Wearable Computers: a Holistic Design Approach
Date and Period: Monday PM, May 5, 2014
Speakers: Roozbeh Jafari, Hassan Ghasemzadeh
Room: Gamma 2
Abstract:
Wearable computers bring to fruition many opportunities to continuously monitor human body with sensors placed on body. They provide new avenues to continuously monitor individuals, whether it is intended to detect an early onset of a disease or to assess the effectiveness of the treatment. In the past few years, the community has observed a large number of applications that have been developed using wearable computers. Yet, not many have been deployed in a large scale. There are still several challenges that need to be addressed before realizing the ubiquitous use of wearable computers.
In this tutorial, we present a review of wearable computing systems and their applications. We highlight several components of wearable computers including signal processing, software, hardware architectures, sensors and actuators. We will in particular review signal processing techniques suitable for time-series data acquired from wearable sensors namely dynamic time warping and hidden Markov models. We provide a review for existing software platforms for wearable computers along with the current state-of-the-art hardware architectures. We emphasize the need for holistic approaches optimizing and enhancing the performance of wearable computers (e.g., reducing their form). We will review several cross-layered techniques aimed at creating hardware accelerators for wearable computers. We will conclude the talk by highlighting opportunities and future directions in wearable computer design.
Roozbeh Jafari (www.essp.utdallas.edu) is an associate professor at UT-Dallas. He received his PhD in Computer Science (UCLA) and completed a postdoctoral fellowship at UC-Berkeley. His research interest lies in the area of wearable computer design and signal processing. His research has been funded by the NSF, NIH, DoD - TATRC, AFRL, AFOSR, DARPA, SRC and industry (Texas Instruments, Tektronix, Samsung, Telecom Italia & Texas Health Resources). He has published over 100 papers in refereed journals and conferences. He has served as the general and technical program committee chairs for several flagship conferences in the area of Wireless Health and Wearable Computers including the ACM Wireless Health 2012 and 2013, International Conference on Body Sensor Networks 2011 and International Conference on Body Area Networks 2011. He is an associate editor for the IEEE Sensors Journal and the IEEE Internet of Things Journal. He is the recipient of the NSF CAREER award (2012) and the RTAS 2011 best paper award.Hassan Ghasemzadeh is currently an Assistant Professor of Electrical Engineering and Computer Science at Washington State University (WSU). His research interests lie in different aspects of Embedded System Design including sustainable and green computing, low-power architectures, reconfigurable computing, and system-level optimization. The focus of his current work is on collaborative signal and information processing, power optimization, data analytics, and algorithm design for networked embedded systems with a primary emphasis on applications in healthcare and wellness. His research spans the areas of embedded systems, computer architecture, signal processing, and machine learning, where he has published more than 80 technical papers on the subject. Prior to joining WSU, he was a Research Manager at the UCLA Wireless Health Institute and an Adjunct Professor of Biomedical Informatics at San Diego State University. He was on the faculty of Azad University from 2003-2006 where he served as Chair of Computer Engineering Department at Damavand branch, Tehran, Iran. Dr. Ghasemzadeh received his Ph.D. degree in Computer Engineering from the University of Texas at Dallas in 2010, and spent the academic year 2010-2011 as a Postdoctoral Fellow at the West Health Institute. He received his M.S. degree in Computer Engineering from University of Tehran, Tehran, Iran, in 2001 and his B.S. degree in Computer Engineering from Sharif University of Technology, Tehran, Iran in 1998.