Mathematics, Statistics, and Computer Science Department &

Hewlett-Packard Present

 

Software Defined Networking:  Next Generation Networking

June 5, 2014

1:00 PM – 4:00 PM (CDT)

Marquette University

Cudahy Hall 001

13th and Wisconsin Ave

(Map)

 

Please join us for a focused discussion and demonstrations of Software Defined Networking (SDN).   SDN, stems from research started seven years ago at Stanford University to deliver easier to manage, scalable, highly orchestrated networks. 

 

We will define the technology, discuss the history, and explore current positioning in networks.   A demonstration will feature creating production ready solutions. We will discuss next-generation research occurring around the world and in Wisconsin. Afterwards, you are welcome to stay and discuss with the presenters and your contemporaries.

 

Program Schedule:

1:00pm:                Definition, History, and Current Technology – Jeff Szczerbinski, CCIE #4052 Emeritus

2:00pm:                Demonstration – David Spears

2:45pm:                Next Generation Research – Aaron Gember – Researcher – UW Madison

3:00pm:                Open discussion

 

Topics for discussion:

VeriCon

VeriCon symbolically reasons about (potentially infinite) network states to verify that network-wide invariants are preserved for any sequence of "events" (e.g., the controller receives a packet header from a switch, or a link fails) and on all admissible topologies, where invariants and topologies are expressed via first-order logic. VeriCon is sound in the sense that if it outputs "no errors" then the preservation of the specified invariants is guaranteed. When verification fails, VeriCon displays a concrete scenario that violates the invariants, in the form of an admissible topology and an event, and it is therefore a useful tool for debugging controller programs.

Learn more at http://agember.com/go/vericon

 

Pratyaastha

Current state of the art distributed SDN controller architectures are insufficient for satisfying tight latency requirements, and they incur more than optimal operating costs. In contrast, Pratyaastha is an efficient and elastic distributed SDN control plane that jointly optimizes both latency and cost. Pratyaastha uses a bipartite graph to reason about the dependencies between SDN application state and SDN switches and find a suitable partitioning of application state. The application state partitions and switches are then assigned to controller instances using a heuristic that solves a variant of multi-dimensional bin packing where the dimensions include both the CPU and memory requirements of the SDN applications. Initial evaluations show that Pratyaastha can yield a 42% reduction in SDN controller operating costs and a 44% decrease in flow-setup latencies.

Learn more at http://agember.com/go/pratyaastha

 

OpenNF

Network functions virtualization (NFV) and software defined networking (SDN) together have the potential to help operators achieve three important goals: (1) offer and satisfy tight service level agreements (SLAs); (2) accurately monitor and manipulate network traffic; and (3) minimize operating expenses. However, operators need additional control mechanisms to be able to satisfy these goals in scenarios where packet processing must be redistributed across a collection of network function instances: e.g., elastic network function scaling, rapid network function upgrades, and selective invocation of advanced remote processing. OpenNF provides efficient, coordinated control of both internal network function state and network forwarding state to allow quick and safe reallocation of flows across network function instances. We address three major challenges in our control plane design: (1) dealing with race conditions, (2) bounding overhead, and (3) accommodating a variety of network functions with minimal changes. Evaluations show that OpenNF can eliminate spurious alerts from intrusion detection systems (IDSs) following scale-out and cut network function scale-in time by tens of minutes compared to using current control frameworks

Learn more at http://agember.com/go/opennf

 

Speakers

 

Jeff Szczerbinski, CCIE #4052 Emeritus, is a veteran in the networking industry. His network career began at the University of Wisconsin Milwaukee back when technologies like DNS did not exist. He has held engineering positions for large enterprises, worked closely with Value Added Resellers and hardware and software manufacturers. Jeff was employed with major technology companies such as Cisco Systems, Polycom, 3COM, and currently Hewlett Packard.  Jeff Szczerbinski holds, CCIE #4052 Emeritus and manages a technology blog around networking which can be found at: www.thenetworkmonkey.com

 

Aaron Gember is a PhD student in the department of Computer Sciences at the University of Wisconsin-Madison. Aaron is a recipient of the IBM PhD Fellowship. Aaron received a B.S. in Computer Science from Marquette University in 2009. His research interests include management abstractions for middleboxes, verification and performance improvement for software-defined networks, and cloud usage patterns of web services.  Aaron also has a strong interested in teaching, which was inspired by his undergraduate research experiences with Embedded Xinu.

 

David Spears holds a degree from MSOE in Electrical Engineering and Controls.  He has a lengthy history in the networking field and has held engineering & leadership positions at Rockwell, Berbee, RW Baird, and Hewlett Packard.  David has a passion for OpenFlow, SDN, and NFV. David was Director of Infrastructure of Infrastructure Services at RW Baird and is currently a Senior Solutions Architect at Hewlett Packard. 

http://h17007.www1.hp.com/us/en/networking/solutions/technology/sdn/index.aspx

https://www.opennetworking.org/

https://www.opennetworking.org/images/stories/downloads/sdn-resources/solution-briefs/sb-sdn-nvf-solution.pdf