Data center bridging

Abstract: Data Center Networks represent the convergence of computing and networking, of data and storage networks, and of packet transport mechanisms in Layers 2 and 3. Congestion control algorithms are a key component of data transport in this type of network. Recently, a Layer 2 congestion management algorithm, called QCN (Quantized Congestion Notification), has been adopted for the IEEE 802.1 Data Center Bridging standard: IEEE 802.1Qau. The QCN algorithm has been designed to be stable, responsive, and simple to implement. However, it does not provide weighted fairness, where the weights can be set by the operator on a per-flow or per-class basis. Such a feature can be very useful in multi-tenanted Cloud Computing and Data Center environments. This paper addresses this issue. Specifically, we develop an algorithm, called AF-QCN (for Approximately Fair QCN), which ensures a faster convergence to fairness than QCN, maintains this fairness at fine-grained time scales, and provides programmable weighted fair bandwidth shares to flows/flow-classes. It combines the QCN algorithm developed by some of the authors of this paper, and the AFD algorithm previously developed by Pan et. al. AF-QCN requires no modifications to a QCN source (Reaction Point) and introduces a very light-weight addition to a QCNcapable switch (Congestion Point). The results obtained through simulations and an FPGA implementation on a 1Gbps platform show that AF-QCN retains the good congestion management performance of QCN while achieving rapid and programmable (approximate) weighted fairness.

Abstract: In an effort to drive down the cost of ownership of interconnection networks for data centers and high-performance computing systems, technologies enabling consolidation of existing networking infrastructure, which often comprises multiple, incompatible networks operating in parallel, are feverishly being developed. Such consolidation carries the promises of lower complexity, less maintenance overhead, and higher efficiency, which should translate into lower power consumption.10-Gigabit Ethernet is one of the contending technologies to fulfill the role of universal data-center interconnect. One of the key features missing from conventional Ethernet is congestion management; this void is being filled by the standardization work of the IEEE 802.1Qau working group. Here, we build on the congestion management scheme defined in 802.1Qau to exploit a crucial property of many data-center networks, namely, multi-pathing, i.e., the presence of multiple alternative paths between any pair of end nodes. We adopt a two-tier approach: in response to congestion detection, an attempt is made to reroute "hot'' flows (i.e., those detected as contributing to congestion) onto an alternative, uncongested path; only when no uncongested alternative exists are transmission rates of hot flows reduced at the sources. We demonstrate how this can lead to significant performance improvements by taking full advantage of path diversity. Moreover, we highlight the scheme's practical usefulness by showing how it improves the performance of parallel benchmark programs on a realistic network.

Abstract: Data Center Virtualization Fundamentals For many IT organizations, todays greatest challenge is to drive more value, efficiency, and utilization from data centers. Virtualization is the best way to meet this challenge. Data Center Virtualization Fundamentals brings together the comprehensive knowledge Cisco professionals need to apply virtualization throughout their data center environments. Leading data center expert Gustavo A. A. Santana thoroughly explores all components of an end-to-end data center virtualization solution, including networking, storage, servers, operating systems, application optimization, and security. Rather than focusing on a single product or technology, he explores product capabilities as interoperable design tools that can be combined and integrated with other solutions, including VMware vSphere. With the authors guidance, youll learn how to define and implement highly-efficient architectures for new, expanded, or retrofit data center projects. By doing so, you can deliver agile application provisioning without purchasing unnecessary infrastructure, and establish a strong foundation for new cloud computing and IT-as-a-service initiatives. Throughout, Santana illuminates key theoretical concepts through realistic use cases, real-world designs, illustrative configuration examples, and verification outputs. Appendixes provide valuable reference information, including relevant Cisco data center products and CLI principles for IOS and NX-OS. With this approach, Data Center Virtualization Fundamentals will be an indispensable resource for anyone preparing for the CCNA Data Center, CCNP Data Center, or CCIE Data Center certification exams. Gustavo A. A. Santana, CCIENo. 8806, is a Cisco Technical Solutions Architect working in enterprise and service provider data center projects that require deep integration across technology areas such as networking, application optimization, storage, and servers. He has more than 15 years of data center experience, and has led and coordinated a team of specialized Cisco engineers in Brazil. He holds two CCIE certifications (Routing & Switching and Storage Networking), and is a VMware Certified Professional (VCP) and SNIA Certified Storage Networking Expert (SCSN-E). A frequent speaker at Cisco and data center industry events, he blogs on data center virtualization at gustavoaasantana.net. Learn how virtualization can transform and improve traditional data center network topologies Understand the key characteristics and value of each data center virtualization technology Walk through key decisions, and transform choices into architecture Smoothly migrate existing data centers toward greater virtualization Burst silos that have traditionally made data centers inefficient Master foundational technologies such as VLANs, VRF, and virtual contexts Use virtual PortChannel and FabricPath to overcome the limits of STP Optimize cabling and network management with fabric extender (FEX) virtualized chassis Extend Layer 2 domains to distant data center sites using MPLS and Overlay Transport Virtualization (OTV) Use VSANs to overcome Fibre Channel fabric challenges Improve SAN data protection, environment isolation, and scalability Consolidate I/O through Data Center Bridging and FCoE Use virtualization to radically simplify server environments Create server profiles that streamline bare metal server provisioning Transcend the rack through virtualized networking based on Nexus 1000V and VM-FEX Leverage opportunities to deploy virtual network services more efficiently Evolve data center virtualization toward full-fledged private clouds -Reviews - The variety of material that Gustavo covers in this work would appeal to anyone responsible for Data Centers today. His grasp of virtualization technologies and ability to relate it in both technical and non-technical terms makes for compelling reading. This is not your ordinary tech manual. Through use of relatable visual cues, Gustavo provides information that is easily recalled on the subject of virtualization, reaching across Subject Matter Expertise domains. Whether you consider yourself well-versed or a novice on the topic, working in large or small environments, this work will provide a clear understanding of the diverse subject of virtualization. -- Bill Dufresne, CCIE 4375, Distinguished Systems Engineer, Cisco (Americas) ..this book is an essential reference and will be valuable asset for potential candidates pursuing their Cisco Data Center certifications. I am confident that in reading this book, individuals will inevitably gain extensive knowledge and hands-on experience during their certification preparations. If youre looking for a truly comprehensive guide to virtualization, this is the one! -- Yusuf Bhaiji, Senior Manager, Expert Certifications (CCIE, CCDE, CCAr), [email protected] When one first looks at those classic Cisco Data Center blueprints, it is very common to become distracted with the overwhelming number of pieces and linkages. By creating a solid theoretical foundation and providing rich sets of companion examples to illustrate each concept, Gustavos book brings hope back to IT Professionals from different areas of expertise. Apparently complex topics are demystified and the insertion of products, mechanisms, protocols and technologies in the overall Data Center Architecture is clearly explained, thus enabling you to achieve robust designs and successful deployments. A must read... Definitely! -- Alexandre M. S. P. Moraes, Consulting Systems Engineer Author of Cisco Firewalls

Abstract: Data center networks today face exciting new challenges in supporting cloud computing and other data-intensive applications. In conventional DCNs, different types of traffic are carried by different types of networks, such as Ethernet and Fibre Channel. Typically, Ethernet carries data traffic among servers in LANs, and Fibre Channel connects servers and storages in storage area networks. Due to the existence of multiple networks, the network cost, power consumption, wiring complexity, and management overhead are often high. The concept of a converged DCN is therefore appealing, carrying both types of traffic in a single converged Ethernet. Recent standards have been proposed for unified data center bridging (DCB) Ethernet and Fibre Channel over Ethernet (FCoE) protocols by the DCB Task Group of IEEE and the T11 Technical Committee of INCITS. In this article, we give a survey of the standards and protocols on converged DCNs, focusing mainly on their motivations and key functionalities. The technologies are discussed mainly from a practical perspective and may serve as a foundation for future research in this area.