Title: The Massachusetts Green High Performance Computing Center
Abstract: The MGHPCC is a high performance computing datacenter built to support the research computing needs of five Boston area universities: MIT, BU, Northeaster, Harvard and UMass. The center is a public/private venture and was constructed with funds from the five universities, the Commonwealth of Massachusetts, Cisco and EMC. This talk present the major infrastructure components of the facility ad explain some of the challenges operating a modern, energy-efficient, high-density computer facility.
Presenter: Jim is the Director of IT Services at the Massachusetts Green High Performance Computing Center. The center is a 10MW computer facility located in Holyoke Massachusetts, an ideal location for computing in Massachusetts with ample land, inexpensive power and access to several high-capacity optical networks. Jim has led the effort to make sure the facility meets the computing and security requirements of the member institutions as the project has progressed through the design, construction, startup and operations phases.
Prior to MGHPCC Jim worked as a consultant to British Telecommunications, CTO at Weather Services International and CTO at MetraTech Corporation. Jim has over 20 years of industry experience as a technology executive, project lead and, operations manager.
Title: Intro to OpenStack Software Platform and Community
Abstract: "The talk introduces the audience to OpenStack, the fastest growing open source project to date, and the preeminent open source cloud operating system. The talk will provide an overview of OpenStack, the various components involved in running a cloud, and how the various components communicate amongst themselves. We describe architectural considerations in designing OpenStack that allow it to be extensible and allow for rapidly scalable development processes. As the OpenStack community has exploded in size, certain processes and procedures were put in place in order to allow for more effective governance and management of the myriad projects that comprise the OpenStack ecosystem. Tooling was developed around blueprint and design development, source code management, code reviews, and team communication, all with the goal of providing more consistent and transparent management of project direction. However, with the introduction of these many processes and tools came a burden to new contributors as well as a need for mentorship programs to nurture and guide individuals and organizations into being good OpenStack community citizens. This talk also provides a guide to highlight these processes and tools to help new folks get introduced to the community. We will be covering the OpenStack community and governance tools we use in the community Review and code best practices, Tips and techniques for best interacting with community members etc."
Bio:
Amrith Kumar is an active technical contributor to the OpenStack project, and a member of the Trove core review team. He is also a member of the OpenStack Foundation Job Analysis Task Force and an author of the book on OpenStack Trove. He is the author of the book on OpenStack Trove (published by Apress, http://www.apress.com/9781484212226). He is the founder and CTO of Tesora, and brings more than two decades of experience delivering industry-leading products for companies specializing in enterprise storage applications, fault tolerant high performance systems and massively parallel databases to Tesora, which he co-founded. Earlier, he served as vice president of technology and product management at Dataupia, maker of the Satori Data Warehousing platform , and Sepaton’s director and general manager where he was responsible for the development of the core virtual tape library product. As a director of product development at Netezza, he managed end-to-end product delivery for all customers and prospects. Amrith studied mathematics at the University of Madras (India) and management at the Indian Institute of Management.
Davanum Srinivas is a long time contributor to OpenStack, currently serves as the Project Technical Lead for Oslo. He is active in several other OpenStack projects like Nova, KeyStone, Magnum and helps with the OpenStack Release team as well. At Mirantis, he leads a team working on improving KeyStone project and infrastructure components like RabbitMQ, MySQL Galera through Oslo projects like Oslo.Messaging and Oslo.DB.
Jonathan Appavoo described project Kittyhawk
Abstract: Kittyhawk represents our vision for a Web-scale computational resource that can accommodate a significant fraction of the world’s computation needs and enable various parties to compete and cooperate in the provisioning of services on a consolidated platform. In this paper, we explain both the vision and the system architecture that supports it. We demonstrate these ideas by way of a prototype implementation that uses the IBM Blue Genet/P platform. In the Kittyhawk prototype, we define a set of basic services that enable the allocation and interconnection of computing resources. By using examples, we show how higher layers of services can be built by using our basic services and standard open-source software.
Abstract:Summary of cloud industry including public cloud space of AWS, Google Cloud, Microsoft Azure, Rackspace; private cloud space: VMware vSphere, OpenStack. Who are the Users of the cloud? Private cloud networking: VMware NSX, Cisco ACI, Openstack networking. Competitive landscpe of private cloud networking: Juniper Conntrail, Midokura, Plumgrid, Dragonflow, Nuage. Technical architecture, pros and cons of each system. Things in the future: bridging between multiple clouds, compatibility with carriers, technical features in the future.
Bio: Richard Wong is a Technical Sales Engineer at Midokura, the open-source virtual networking company a leading contributor in the Neutron and Kuryr networking projects. He has deep systems engineering experience, with a focus on systems architecture, network design, and datacenter planning. Richard has worked in a wide variety of platform environments, including VMware, OpenStack, RHEL, Solaris, Windows, AIX, Ubuntu, and has a wide background in SAN storage as well as Cisco networking. He is native of a New York City and have worked at firms such as Cantor Fitzgerald, AIG, DTCC, Dell, Egenera.
Title: Unpacking the box: The evolution of both the cloud data center network and Big Switch Networks, too
Abstract: The requirements placed on the network within the data center have changed rapidly over the last ten years, with solutions evolving to L2/L3 SDN fabrics. A quick survey of the factors that drove the change, the technology that enables hyperscale architecture solutions, the advantages SDN fabrics have and Big Switch Networks' journey through it all.
Bio: Paul Zablotski is a Systems Engineer at Big Switch Networks. He has more than 20 years' experience in networking and compute, including previous roles at Juniper Networks, Enterasys Networks, Dunkin' Brands and Fore Systems/Marconi Communications. He's attained many IT industry certifications, though none is more prized than JOAT-MOS.